MPSDS JPSM Seminar Series

November 30, 2022

12:00 - 1:00 EST

Effect of Branching Middle Responses in Dichotomous Polar Scales in Web Surveys

In telephone surveys, 11 to 49% of respondents would select a middle alternative when it is offered although they would not volunteer it if it were not mentioned in dichotomous bipolar questions. Furthermore, offering a middle option led to differences in response effects that are related to respondent characteristics, including social desirability bias and satisficing effects. While a question form that branches middle responses has been shown to have a lower validity compared to offered form in telephone surveys, potentially, branched question form can motivate respondents to spend extra time and effort in giving a response in the absence of an interviewer. Therefore, differences in validity and reliability of responses to branched question form compared to offered form is a research interest in general population web surveys. This study tests the validity and the reliability to branched question form in a general population survey using a randomized experiment. The branched question form did not change validity and reliability of responses and reduced the satisficing behavior based on the proxies compared to the offered form.

Z. Tuba Suzer Gurtekin is an Assistant Research Scientist within the Institute for Social Research (ISR) at the University of Michigan. She is the scientific leader of the Surveys of Consumers, which conducts monthly national surveys of American households to understand consumer expectations and how those expectations impact their spending and saving behavior. Her research experience has included development of alternative sample, methodology and questionnaire designs, data collection and analysis methods for a general population in parallel survey modes. In addition to her work through the Surveys of Consumers, she also currently serves on the Board of Associate Editors of CDC’s Preventing Chronic Disease Journal. 608.82She teaches survey sampling and survey methodology in University of Michigan’s Clinical Research Design and Statistical Analysis Program (OJOC CRDSA).

MPSDS

The University of Michigan Program in Survey Methodology was established in 2001 seeking to train future generations of survey and data scientists. In 2021, we changed our name to the Michigan Program in Survey and Data Science. Our curriculum is concerned with a broad set of data sources including survey data, but also including social media posts, sensor data, and administrative records, as well as analytic methods for working with these new data sources. And we bring to data science a focus on data quality — which is not at the center of traditional data science. The new name speaks to what we teach and work on at the intersection of social research and data. The program offers doctorate and master of science degrees and a certificate through the University of Michigan. The program's home is the Institute for Social Research, the world's largest academically-based social science research institute.

SISRT

June 5 – July 28, 2023

The Summer Institute in Survey Research Techniques is a teaching program of the Survey Research Center at the Institute for Social Research. It is located on the central campus of the University of Michigan at 426 Thompson Street in Ann Arbor. The summer courses are select offerings from the Michigan Program in Survey and Data Science, and can be used to pursue a doctorate, master of science and a certificate in survey methodology.

All 2023 courses will be offered in an alternative remote format with the exception of the Sampling Program for Survey Statisticians. Payment of Summer Scholar and workshop fees must be made in full before you will be officially registered for class. Fees are based on total “course hours” (assigned to each course as shown in the section on description of courses and on the 2023 course schedule) although no formal academic credit is actually earned.

Our courses this summer will be offered primarily by two-way, live video through a platform that supports lectures and group work. In some cases, courses are offered in a flipped format in which lectures are video recorded for students to watch on-demand and then meet with their instructor by two-way live video to discuss the lectures, readings, and problem sets. All classes are scheduled in Eastern Standard Time Zone.

We have been offering courses in remote formats for many years through our connection with the graduate programs at the Universities of Michigan and Maryland which share all courses by live classroom-to-classroom video. In the COVID era, our transition to entirely remote instruction has been straightforward and brought the students’ experience very close to that of a place-based classroom.

We understand that some participants were looking forward to visiting Ann Arbor, networking and participating in social activities. As an alternative, we are planning several virtual social and networking activities in which participants will meet informally (by live video) with their instructors and just with each other in small groups to discuss various topics, some related to courses and some not. This will give participants a chance get to know each other as well as instructors outside the “classroom.” We’re excited to work with you as we learn how to best connect with each other remotely.

Abstract:

CRISPR genome editing technologies and single-cell assays have opened new opportunities to study cellular systems and gene regulation at an unprecedented level of detail.

In this talk, I will first present computational methods we have developed to uncover and dissect regulatory elements using CRISPR genome editing technologies. I will also discuss challenges associated with using CRISPR technologies related to designing perturbations and quantifying editing outcomes.

I will then cover our work in modeling data from current single-cell assays, discussing methods to uncover development trajectories, recover RNA-velocity with uncertainty, and create interpretable regulatory maps from multi-omics data using graph embedding techniques.

https://umich-health.zoom.us/j/93929606089?pwd=SHh6R1FOQm8xMThRemdxTEFMWWpVdz09

Short bio:

Luca Pinello is a computational biologist and leader in developing computational methods for functional genomics, genome editing and single cell technologies. He holds a Ph.D. in Mathematics and Computer Science from University of Palermo, Italy. He is currently an Associate Pathologist at Massachusetts General Hospital (MGH) and an Associate Professor of Pathology at Harvard Medical School. He is also part of the MGH Center for Cancer Research and an Associate Member of the BROAD Institute of MIT and Harvard. He has developed several foundational computational tools in the field of genome editing for the design (CRISPRme, CRISPRitz, PrimeDesign), quantification (CRISPResso 1 and 2), and analyses of coding and non-coding tiling screens (CRISPRO, CRISPR-SURF). He was awarded one of the first NIH R35 Genomic Innovator Awards, a prestigious grant supporting highly innovative researchers working on important problems in genomics.

CGIS offers First Steps sessions virtually (via Zoom) every Monday and Thursday from 4:00pm to 4:30pm during the academic year while classes are in session, with the exception of holidays.

First Step sessions are a great opportunity to learn more about the application process prior to meeting with an advisor. You can learn about all of our programs around the world, scholarships and other financial aid resources, the CGIS application process, and more!

*Attending a First Step session is no longer a required component of the CGIS application process.*

CGIS offers First Steps sessions virtually (via Zoom) every Monday and Thursday from 4:00pm to 4:30pm during the academic year while classes are in session, with the exception of holidays.

First Step sessions are a great opportunity to learn more about the application process prior to meeting with an advisor. You can learn about all of our programs around the world, scholarships and other financial aid resources, the CGIS application process, and more!

*Attending a First Step session is no longer a required component of the CGIS application process.*

CGIS offers First Steps sessions virtually (via Zoom) every Monday and Thursday from 4:00pm to 4:30pm during the academic year while classes are in session, with the exception of holidays.

First Step sessions are a great opportunity to learn more about the application process prior to meeting with an advisor. You can learn about all of our programs around the world, scholarships and other financial aid resources, the CGIS application process, and more!

*Attending a First Step session is no longer a required component of the CGIS application process.*

First Step sessions are a great opportunity to learn more about the application process prior to meeting with an advisor. You can learn about all of our programs around the world, scholarships and other financial aid resources, the CGIS application process, and more!

*Attending a First Step session is no longer a required component of the CGIS application process.*

This talk is an overview of the topic of this semester's learning seminar. It is also a planning meeting as we will also assign speakers to talks.

]]>

First Step sessions are a great opportunity to learn more about the application process prior to meeting with an advisor. You can learn about all of our programs around the world, scholarships and other financial aid resources, the CGIS application process, and more!

*Attending a First Step session is no longer a required component of the CGIS application process.*

Title: Tate Classes and Endoscopy for GSp4

Abstract: Weissauer proved using the theory of endoscopy that the Galois representations associated to classical modular forms of weight two appear in the middle cohomology of both a modular curve and a Siegel modular threefold. Correspondingly, there are large families of Tate classes on the product of these two Shimura varieties, and it is natural to ask whether one can construct algebraic cycles giving rise to these Tate classes. It turns out that a natural algebraic cycle generates some, but not all, of the Tate classes: to be precise, it generates exactly the Tate classes which are associated to generic members of the endoscopic L-packets on GSp4. In the non-generic case, one can at least show that all the Tate classes arise from Hodge cycles. I'll explain these results and sketch their proofs, which rely on the theta correspondence.

We will schedule talks for the rest of the semester. Suggestions for topics that you would like to hear about and topics you would like to speak about are welcome!

]]>A category central to algebraic combinatorics is that of “rings with bases”. We’ll look at a very simple example, the nil Hecke algebra, and its action on polynomials in infinitely many variables. The resulting dual basis is the “Schubert polynomials”. Surprisingly, these have positive coefficients, which we will compute by counting “pipe dreams”.

]]>Everyone who would like to give a talk or would like to suggest a topic for a talk is encouraged to come!

]]>This talk introduces a mean field game for a family of filtering problems related to the classic sequential testing of a Brownian motion’s drift. In our formulation, agents observe a private signal process and want to make a determination about an unknown binary state of nature. The game arises by allowing the drift of the signal process to incorporate information about the other agents’ actions and enforcing that each agent must minimize an associated Bayes risk. In this setting we are able to develop a deep understanding of the solution structure, establish the existence of a mean field equilibrium, and study the equilibria numerically. To the best of our knowledge, this work presents the first treatment in the literature of a tractable mean field game with information filtering, optimal stopping, and a common unobserved noise. This presentation is based on recent joint work with Yuchong Zhang at the University of Toronto.

]]>The Mozes-Shah theorem classifies limits of measures which are invariant under unipotent flows. This is the most common situation that arises in applications of measure classification in Number Theory and Geometry.

I will show some cool examples of this theorem and explain the main steps in the proof of the theorem (relying on the linearization technique of Dani-Margulis).

It will be an RTG talk, mainly focusing on examples.

If time permits I will explain a quantitative version of the theorem due to Einsiedler-Margulis-Venkatesh.

First Step sessions are a great opportunity to learn more about the application process prior to meeting with an advisor. You can learn about all of our programs around the world, scholarships and other financial aid resources, the CGIS application process, and more!

*Attending a First Step session is no longer a required component of the CGIS application process.*

TBA

]]>In this talk, we discuss a conical extension of averaged non-expansive operators and its role in convergence analysis of fixed point algorithms. In particular, we study various properties, for example, stability under relaxations, convex combinations and compositions of conically averaged operators. We then utilize such properties in order to analyze the convergence of proximal point algorithm, forward-backward algorithm, and the adaptive Douglas-Rachford algorithm. This talk is based on joint work with Sedi Bartz and Minh Dao.

Time: Friday Jan 13, 09:00 AM- 10:00 AM, Eastern Time (US and Canada)

Zoom Link: https://umich.zoom.us/j/92332350184?pwd=c1hJZmRlcGV5VVJWRTJwRDhTdTFVZz09

Meeting ID: 923 3235 0184

Passcode: 123456

The Seminar organization: Anthony Bloch, Boris Mordukhovich, Nguyen-Truc-Dao Nguyen

We will plan the meetings for the rest of the semester. Please come with a few topics you would be interested in hearing or presenting on.

]]>Putting hyperbolic metrics on a finite-type surface S gives us linear representations of the fundamental group of S into PSL(2,R) with many nice geometric and dynamical properties: for instance they are discrete and faithful, and in fact stably quasi-isometrically embedded.

In this talk, we will introduce (relatively) Anosov representations, which generalise this picture to higher-rank Lie groups such as PSL(d,R) for d>2, giving us a class of (relatively) hyperbolic subgroups there with similarly good geometric and dynamical properties.

This is mostly joint work with Andrew Zimmer.

N/A

]]>Josh Cassada will answer questions live from the International Space Station, and commentary will be provided by former NASA engineer John Foster, U-M Applied Physics Alumnus and Professor of Nuclear Engineering and Radiological Sciences and Professor of Aerospace Engineering, College of Engineering.

Click this URL to send us a question for Josh: https://myumi.ch/Ek1Xr

Please submit your question by 1/11/23. Those in attendance may have the opportunity to read their question during the event.

Discover more about Josh Cassada and this event: http://www.saturdaymorningphysics.org

Event will also be available via live stream on YouTube: https://youtu.be/34tEqQyB1Mk

Only this SMP event will be located in a different venue: 1420 Central Campus Classroom Bldg., 1225 Geddes, Ann Arbor, MI

Let f be a dominant polynomial transformation of the complex affine plane. The dynamical degree lambda_1 of f is defined as the limit of the n-th root of the degree of the n-th iterate of f. In 2007, Favre and Jonsson showed that the dynamical degree of any polynomial endomorphism of the affine plane is a quadratic integer. For any affine surface S0, there is a definition of the dynamical degree that generalizes the one on the affine plane. We show that the result still holds for any complex affine surface: the dynamical degree of an endomorphism of any complex affine surface is a quadratic integer. The proof uses the space of valuations centered at infinity V. The endomorphism f defines a transformation of V and studying the dynamics of f on V gives information about the dynamics of f on S0. The main result is that under certain hypothesis, f admits an attracting fixed point in V that we call an eigenvaluation. This implies that one can find a good compactification S of S0 such that f admits an attracting fixed point p at infinity and f has a normal form at p; the result on the dynamical degree follows from the normal form.

]]>In a graded ring, a homogeneous ideal is an ideal which is generated by homogeneous elements. This seems straightforward enough, but if your ideal is presented abstractly instead of in terms of generators, it becomes less obvious how to verify homogeneity. In this talk, we will discuss several different techniques for showing an ideal is homogeneous, as well as comparing the strengths and limitations of these techniques. We'll illustrate these via proving useful results on how homogeneity is preserved, such as showing that associated primes of homogeneous ideals are homogeneous ideals, and showing the integral closure of a homogeneous ideal is homogeneous.

]]>Please join us for our planning meeting! We will discuss what topics we want to present/hear this semester. There will be a snack.

Note the unusual time/place.

The phenomenon of wave propagation in random environments appears in many physical situations of practical interest. The simplest model for this phenomenon is the Schrodinger equation coupled to a weak random potential, which describes the evolution of an electron in disordered media. The effect of the disorder is to scatter the wave into random directions. The long-time behavior is described by an effective diffusion equation, which was first established by Erdös, Salmhofer, and Yau using sophisticated diagrammatic arguments. In this talk I will describe a new approach to proving this effective limit which uses a wavepacket decomposition of the solution to give a geometric meaning to the diagrams. I will focus on the geometry of the diagrams and state some elementary open problems concerning Euclidean geometry which suggest a path to simpler proofs and stronger results.

Zoom Room: https://umich.zoom.us/j/96021646996

We start with the Bruhat decomposition of the full flag variety and the Borel presentation of its cohomology ring to introduce the Schubert polynomials, the main objects of interest for this seminar. We will then briefly discuss several topics that can be covered throughout the semester, including various monomial expansions of the Schubert polynomials, combinatorial algebraic geometry of the Schubert varieties, Gröbner geometry of matrix Schubert varieties and more.

]]>The Mordell--Weil Theorem states that the rational points of an abelian variety are finitely generated. The proof is in two steps: first, we will reduce the problem to the weak Mordell--Weil Theorem using the theory of heights. Then we will prove the weak Mordell--Weil Theorem from the finiteness of the Selmer group. The talk will be accessible to anyone who has seen some algebraic number theory.

]]>We present results on Reflected Backward Stochastic Differential Equations (RBSDE for short) on Brownian filtration with two optional barriers satisfying some separation condition.

It has been widely recognized that RBSDEs provide a useful framework for studying problems in many fields, such as financial mathematics, stochastic optimal control and partial differential equations (e.g. optimal stopping problem, Dynkin games, stopping and control games, switching problem, PDEs with singular data, homogenization, boundary problems, regularity problems, numerical schemes etc.).

The theory of RBSDEs is well studied for càdlàg barriers. However, there is only a few papers concerning BSDEs with non-càdlàg barriers. It is caused mostly by the following: the main component of the solution does not have to be a càdlàg process, the minimality condition (guaranteeing uniqueness) is complicated and unintuitive, and the basic proof technique of càdlàg case, i.e. penalization method, does not apply to non-càdlàg case.

We will present the results on the existence and uniqueness of a solution to RBSDEs with two optional barriers. The generator is assumed to be non-increasing with respect to the value variable (with no restrictions on the growth) and Lipschitz continuous, with sublinear growth, with respect to the control variable. Data are in Lp for some p≥1. We also present some results concerning methods of approximation of the solution via modified penalization scheme, and present the link between solutions of RBSDEs with optional barriers and the value processes for generalized nonlinear Dynkin games.

The results were obtained in cooperation with Tomasz Klimsiak and Leszek Słomiński.

A metric structure is a metric space equipped with various

functions, relations, and constants. We will introduce metric

structures and signatures, as well as some of the examples to which

continuous logic can be applied.

I will talk about expected utility maximization under ratchet and drawdown constraints on consumption in incomplete semimartingale markets. The drawdown constraint on consumption means that the consumption rate process does not fall below a fraction \lambda\in[0,1] of its current running maximum; ratchet constraint, a special case corresponding to \lambda=1, means that consumption rate is non-decreasing. For each \lambda\in[0,1], the optimization is considered via convex duality methods and with respect to two parameters: the initial wealth and the essential lower bound on consumption process. In order to state the problem and define the primal domains in this general semimartingale market setting, I introduce a natural extension of the notion of running maximum to arbitrary non-negative optional consumption rate processes. The dual domains for optimization are then characterized in terms of the closed solid hull of the set of equivalent martingale deflators with respect to a certain ordering on the set of non-negative optional processes. Finally, I will show that the abstract duality result for incomplete markets can be used in order to derive a more detailed characterization of solutions in the complete market case.

]]>A real form of a complex algebraic variety X is a real algebraic variety whose complexification is isomorphic to X. Many families of complex varieties have a finite number of nonisomorphic real forms, but up until recently no example with infinitely many had been found. In 2018, Lesieutre constructed a projective variety of dimension six with infinitely many nonisomorphic real forms, and this year, Dinh, Oguiso and Yu described projective rational surfaces with infinitely many as well. In this talk, I’ll present the first example of a rational affine surface having uncountably many nonisomorphic real forms.

]]>

First Step sessions are a great opportunity to learn more about the application process prior to meeting with an advisor. You can learn about all of our programs around the world, scholarships and other financial aid resources, the CGIS application process, and more!

*Attending a First Step session is no longer a required component of the CGIS application process.*

Introduction and organizational meeting

]]>MPSDS JPSM Seminar Series

February 1, 2023

12:00 - 1:00 EST

The Role of Data Collection in Population Science: Contemporary Studies from ABCD to HBCD

Abstract

Recently nationwide consortiums of multiple research sites have conducted multi-modal, longitudinal cohort studies and provided unprecedented data sources for population science research. For example, the Adolescent Brain Cognitive Development (ABCD) Study has collected data from 11,880 children ages 9-10 across 21 U.S. research sites, as the largest long-term study of brain development and child health; and the Healthy Brain and Child Development (HBCD) Study will enroll 7,500 pregnant women across 25 research sites and follow them from pregnancy through early childhood, as the largest long-term study of early brain and child development in the U.S. Both studies aim to reflect the sociodemographic diversity of the target population to enable characterization of natural variability and trajectories. Without probability sampling as the touchstone for randomization-based inferences, the data quality and analysis validity require rigorous evaluations and potentially rely on untestable assumptions. The data collection process also presents various challenges during practical operation.

In this talk, I look into both inference and design schemes to study the impact of data collection on population science. First, using the ABCD study as an example of secondary data analysis, I discuss inference approaches focusing on multilevel regression and poststratification for population generalizability and latent subgroup detection for population heterogeneity in brain activity and association studies. Second, I introduce the HBCD study design. HBCD also aims to include individuals demographically and behaviorally similar to those in the substance exposure group, but without exposure, to enable valid causal inference in a non-experimental study design. I discuss our proposed weighting, matching, and modeling strategies to leverage analysis goals to inform the design and dashboard monitoring for adaptive sample enrollment.

Bio

Yajuan Si is a Research Associate Professor in the Institute for Social Research at the University of Michigan. Dr Si’s research lies in cutting-edge methodology development in streams of Bayesian statistics, linking design- and model-based approaches for survey inference, missing data analysis, confidentiality protection involving the creation and analysis of synthetic datasets, and causal inference with observational data.

Michigan Program in Survey and Data Science (MPSDS)

The University of Michigan Program in Survey Methodology was established in 2001 seeking to train future generations of survey and data scientists. In 2021, we changed our name to the Michigan Program in Survey and Data Science. Our curriculum is concerned with a broad set of data sources including survey data, but also including social media posts, sensor data, and administrative records, as well as analytic methods for working with these new data sources. And we bring to data science a focus on data quality — which is not at the center of traditional data science. The new name speaks to what we teach and work on at the intersection of social research and data. The program offers doctorate and master of science degrees and a certificate through the University of Michigan. The program's home is the Institute for Social Research, the world's largest academically-based social science research institute.

Summer Institute in Survey Research Techniques (SISRT)

The mission of the Summer Institute is to provide rigorous and high quality graduate training in all phases of survey research. The program teaches state-of-the-art practice and theory in the design, implementation, and analysis of surveys. The Summer Institute in Survey Research Techniques has presented courses on the sample survey since the summer of 1948, and has offered such courses every summer since. Graduate-level courses through the Program in Survey and Data Science are offered from June 5 through July 28 and available to enroll in as a Summer Scholar.

The Summer Institute uses the sample survey as the basic instrument for the scientific measurement of human activity. It presents sample survey methods in courses designed to meet the educational needs of those specializing in social and behavioral research such as professionals in business, public health, natural resources, law, medicine, nursing, social work, and many other domains of study.

MPSDS JPSM Seminar Series

February 1, 2023

12:00 - 1:00 EST

The Role of Data Collection in Population Science: Contemporary Studies from ABCD to HBCD

Abstract

Recently nationwide consortiums of multiple research sites have conducted multi-modal, longitudinal cohort studies and provided unprecedented data sources for population science research. For example, the Adolescent Brain Cognitive Development (ABCD) Study has collected data from 11,880 children ages 9-10 across 21 U.S. research sites, as the largest long-term study of brain development and child health; and the Healthy Brain and Child Development (HBCD) Study will enroll 7,500 pregnant women across 25 research sites and follow them from pregnancy through early childhood, as the largest long-term study of early brain and child development in the U.S. Both studies aim to reflect the sociodemographic diversity of the target population to enable characterization of natural variability and trajectories. Without probability sampling as the touchstone for randomization-based inferences, the data quality and analysis validity require rigorous evaluations and potentially rely on untestable assumptions. The data collection process also presents various challenges during practical operation.

In this talk, I look into both inference and design schemes to study the impact of data collection on population science. First, using the ABCD study as an example of secondary data analysis, I discuss inference approaches focusing on multilevel regression and poststratification for population generalizability and latent subgroup detection for population heterogeneity in brain activity and association studies. Second, I introduce the HBCD study design. HBCD also aims to include individuals demographically and behaviorally similar to those in the substance exposure group, but without exposure, to enable valid causal inference in a non-experimental study design. I discuss our proposed weighting, matching, and modeling strategies to leverage analysis goals to inform the design and dashboard monitoring for adaptive sample enrollment.

Bio

Yajuan Si is a Research Associate Professor in the Institute for Social Research at the University of Michigan. Dr Si’s research lies in cutting-edge methodology development in streams of Bayesian statistics, linking design- and model-based approaches for survey inference, missing data analysis, confidentiality protection involving the creation and analysis of synthetic datasets, and causal inference with observational data.

Michigan Program in Survey and Data Science (MPSDS)

The University of Michigan Program in Survey Methodology was established in 2001 seeking to train future generations of survey and data scientists. In 2021, we changed our name to the Michigan Program in Survey and Data Science. Our curriculum is concerned with a broad set of data sources including survey data, but also including social media posts, sensor data, and administrative records, as well as analytic methods for working with these new data sources. And we bring to data science a focus on data quality — which is not at the center of traditional data science. The new name speaks to what we teach and work on at the intersection of social research and data. The program offers doctorate and master of science degrees and a certificate through the University of Michigan. The program's home is the Institute for Social Research, the world's largest academically-based social science research institute.

Summer Institute in Survey Research Techniques (SISRT)

The mission of the Summer Institute is to provide rigorous and high quality graduate training in all phases of survey research. The program teaches state-of-the-art practice and theory in the design, implementation, and analysis of surveys. The Summer Institute in Survey Research Techniques has presented courses on the sample survey since the summer of 1948, and has offered such courses every summer since. Graduate-level courses through the Program in Survey and Data Science are offered from June 5 through July 28 and available to enroll in as a Summer Scholar.

The Summer Institute uses the sample survey as the basic instrument for the scientific measurement of human activity. It presents sample survey methods in courses designed to meet the educational needs of those specializing in social and behavioral research such as professionals in business, public health, natural resources, law, medicine, nursing, social work, and many other domains of study.

Unitary Shimura varieties are moduli spaces of abelian varieties with certain extra structures. A fruitful way to study their geometry is by considering stratifications of the space. One such stratification is the Ekedahl-Oort (E-O) stratification defined with respect to the structure of the p-torsion group scheme. We follow the work of Moonen and Wooding to investigate the interaction of the E-O strata with other aspects of the geometry of the Shimura variety. We present a few novel results and observations in some specific cases, as well as some future directions. Time permitting, we'll also talk about the Newton stratification.

]]>

First Step sessions are a great opportunity to learn more about the application process prior to meeting with an advisor. You can learn about all of our programs around the world, scholarships and other financial aid resources, the CGIS application process, and more!

*Attending a First Step session is no longer a required component of the CGIS application process.*

We study the statistical behavior of quantum entanglement in bipartite systems over fermionic Gaussian states as measured by von Neumann entropy. The formulas of average von Neumann entropy with and without particle number constrains have been recently obtained, whereas the main results of this work are the exact yet explicit formulas of variances for both cases. For the latter case of no particle number constrain, the results resolve a recent conjecture on the corresponding variance. Different than the existing methods in computing variances over other generic state models, proving the results of this work relies on a new simplification framework. The framework consists of a set of new tools in simplifying finite summations of what we refer to as dummy summation and re-summation techniques. As a byproduct, the proposed framework leads to various new transformation formulas of hypergeometric functions. This talk is based on the joint work with Lu Wei available at https://arxiv.org/abs/2211.16709

A recording of the talk can be found here: https://youtu.be/sGdpLRdiorI

Do you have a combinatorics-related picture that you cannot stop thinking about? Bring it next week and share its marvelousness and what it means (or why you want to know what it means) and why you think it's cool -- lightning-talk style!

]]>We may think of degeneration as a way to approximate a given ideal or ring by a possibly simpler ideal or ring. We will begin by understanding degenerations of an ideal in a polynomial ring to a monomial ideal (called Grobner degeneration). Analogously, we will also consider toric degenerations of rings. Finally, we will see some concrete applications of these techniques. Time permitting, we will also discuss different techniques for producing such degenerations.

]]>Tame geometry was suggested by Grothendieck as a possible paradise where the geometer would not have to worry about the existence of pathologies of real analysis. Developed by model theorists as o-minimal geometry, tame geometry has had spectacular applications to algebraic geometry in the last 15 years. This talk will review the key concepts and present some of these applications.

]]>We will be talking about the main ways to compute the Schubert polynomials from permutations, relationships between Schubert polynomials and RC-graphs, chute moves on RC-graphs, and the Monk's rule.

]]>We study the asymptotic behavior of the normalized maxima of real-valued diffusive particles with mean-field drift interaction. Our main result establishes propagation of chaos: in the large population limit, the normalized maxima behave as those arising in an i.i.d system where each particle follows the associated McKean—Vlasov limiting dynamics. This allows for the asymptotic distribution of the normalized maxima to be determined by using results from standard Extreme—Value Theory. The proof uses a change of measure argument that depends on a delicate combinatorial analysis of the iterated stochastic integrals appearing in the chaos expansion of the Radon-Nikodym density. Our work is motivated by problems arising in stochastic portfolio theory, credit risk and mean-field games. Possible extensions to more complex settings are also discussed.

]]>We will introduce the syntax and semantics for continuous

logic, defining the formulas of continuous logic and what it means for

a formula to be true in a metric structure.

We show that given a class of comonotonic claims there is a probability measure Q such that the dynamic spectral risk measure of each such claim is a martingale under Q. The applications explored of this result are: (1) the use of statistical and calibration techniques that are typically employed under the law of one price, such as digital moment estimation and fast Fourier transform; (2) a simplified numerical scheme for the nonlinear valuation of financial claims and of portfolio selection that minimizes the worst case scenario value; (3) the definition of a dynamic, time-consistent, convex, but not coherent spectral risk measure, which allows the introduction of diminishing marginal returns without the theoretical limitations of expected utility theory

]]>

First Step sessions are a great opportunity to learn more about the application process prior to meeting with an advisor. You can learn about all of our programs around the world, scholarships and other financial aid resources, the CGIS application process, and more!

*Attending a First Step session is no longer a required component of the CGIS application process.*

Hodge theory, as developed by Deligne and Griffiths, is one of the main tools for analysing the geometry and arithmetic of complex algebraic varieties, that is, solution sets of algebraic equations over the complex numbers. It associates to any complex algebraic variety an apparently simple linear algebra gadget: a finite dimensional vector space over the rationals, whose complexification is naturally endowed with two filtrations. Hodge theory occupies a central position in mathematics through its relations to differential geometry, algebraic geometry, differential equations and number theory.

It is an essential fact that at heart, Hodge theory is not algebraic but rather the transcendental comparison of two algebraic structures. On the other hand, some of the deepest conjectures in mathematics (the Hodge conjecture and the Grothendieck period conjecture) suggest that this transcendence is severely constrained. In these lectures, we survey the recent advances bounding this transcendence, mainly due to the introduction of tame geometry as a natural framework for Hodge theory.

This is the first of a 2-3 part series on elementary Teichmüller theory, and on the SL(2, R) action on related moduli spaces. We will also explore links between the dynamics on Teichmüller space to the dynamics of discrete subgroups of Lie groups on homogeneous spaces.

]]>Most optimization problems involve uncertain data due to measurement errors, unknown future developments and modeling approximations. Stochastic optimization assumes that the uncertain parameter is probabilistic. An other approach is called robust optimization which expects the uncertain parameter to belong to a set that is known prior. In this talk, we consider scalar optimization problems under uncertainty with infinite scenario sets. We apply methods from vector optimization in general spaces, set-valued optimization and scalarization techniques to derive necessary optimality conditions for solutions of robust optimization problems.

]]>No combinatorics seminar on this date.

]]>A lovely feature of algebraic varieties is that they admit compactifications. Even better, it often happens that the geometry of a compactification is influenced, or even determined, by combinatorial data, as occurs in the correspondence between toric varieties (certain compactifications of tori) and fans (combinatorial data). Log geometry is a powerful formalism for keeping track of combinatorial data at the boundary of a compactification, and since its introduction in the 1980's, it has played an important role in a wide range of topics from p-adic Hodge theory to the birational geometry of moduli spaces. The goal of the talk is to give a brief introduction to the basic language of log geometry, with no previous exposure to toric geometry assumed.

]]>The surface group representations of Euler class zero into PSL(2,R) are those which are deformations of diagonal (reducible) representation. Marche'-Wolff showed that the mapping class group acts ergodically on these representations with the strange exception of genus 2. We explore the infinitesimal data describing a deformation from a diagonal representation to an irreducible representation, and examine how the Torelli group acts on these deformations. Joint with James Farre.

]]>Hodge theory, as developed by Deligne and Griffiths, is one of the main tools for analysing the geometry and arithmetic of complex algebraic varieties, that is, solution sets of algebraic equations over the complex numbers. It associates to any complex algebraic variety an apparently simple linear algebra gadget: a finite dimensional vector space over the rationals, whose complexification is naturally endowed with two filtrations. Hodge theory occupies a central position in mathematics through its relations to differential geometry, algebraic geometry, differential equations and number theory.

It is an essential fact that at heart, Hodge theory is not algebraic but rather the transcendental comparison of two algebraic structures. On the other hand, some of the deepest conjectures in mathematics (the Hodge conjecture and the Grothendieck period conjecture) suggest that this transcendence is severely constrained. In these lectures, we survey the recent advances bounding this transcendence, mainly due to the introduction of tame geometry as a natural framework for Hodge theory.

Are you considering applying for an internship? Hear from your peers regarding their experiences!

This is an informal discussion session with a student panel, consisting of AIM PhD students with prior internship experiences. The panel will provide a overview of their respective journeys, and attendees will have nearly the full duration of the seminar to ask questions.

The RSK algorithm gives a beautiful bijection between the symmetric group S_n and pairs of young tableaux with the same shape. In this interactive talk we will see several ways to implement the algorithm, from patience sorting to growth diagrams to geometric constructions. There will be lots of examples (and maybe a little representation theory).

]]>Title: Global Shimura varieties

]]>

First Step sessions are a great opportunity to learn more about the application process prior to meeting with an advisor. You can learn about all of our programs around the world, scholarships and other financial aid resources, the CGIS application process, and more!

*Attending a First Step session is no longer a required component of the CGIS application process.*

Roughly 15 years ago, Alexander Katsevich, motivated by tomography, began to study the singular values of the finite Hilbert transform (FHT) acting on interval subsets of the real line. This was the beginning of a (ongoing) program that has produced 10+ papers by 5 authors dedicated to the spectral theory of the FHT. We mention some of the history, the connection to tomography and conclude by discussing our latest result, which deals with the diagonalization of the FHT acting on many intervals with touching endpoints.

]]>Title: Geometry and arithmetic of bielliptic Picard curves

Abstract: I'll describe the very pretty geometry of the curves y^3 =

x^4 + ax^2 + b, and use it to "see" the quaternionic multiplication on

their Prym varieties, giving a very explicit family of QM abelian

surfaces (sometimes called "false elliptic curves"). I'll then

describe a few recent results on the arithmetic of these surfaces: a)

a full classification of all rational torsion points in this family

and b) a proof that the average rank in the corresponding family of

Pryms is at most 3. This is based on joint work with Laga and

Laga-Schembri-Voight.

Toric ideals are ideals of polynomial rings with interesting algebraic, geometric, and combinatorial properties. In this talk, we'll introduce toric ideals and give several examples. We'll show how to determine Groebner bases for them, as well as discuss their applications to algebraic statistics. If time permits, we'll also give a combinatorial method for calculating their syzygies and multigraded Betti numbers.

]]>Berkovich spaces are analogues of complex manifolds when the complex numbers are replaced by a non-Archimedean field, that is, a field satifying the strong triangle inequality.

I will discuss two instances where Berkovich spaces naturally appear within complex geometry. The first concerns the Yau--Tian--Donaldson conjecture, on the existence of Kähler--Einstein metrics on Fano manifolds. The second situation appears in the context of degenerations of Calabi--Yau manifolds, and features conjectures by Strominger--Yau--Zaslow, and Kontsevich--Soibelman.

This is based on joint work with R. Berman, S. Boucksom, J, Hultgren, E. Mazzon, and N. McCleerey.

Instead of a talk today, we'll be having snack time! Join us at 5 pm to chat and have some food.

]]>We introduce combinatorial formulas of (double) Schubert and Grothendieck polynomials based on bumpless pipe dreams and give a combinatorial proof of Monk’s rule for Schubert and double Schubert polynomials using bumpless pipe dreams that generalizes Schensted’s insertion on semi-standard Young tableaux. We also give a bijection between pipe dreams and bumpless pipe dreams and discuss its canonical nature.

]]>We study graphon mean-field backward stochastic differential equations (BSDEs) with jumps and associated dynamic risk measures. We establish the existence, uniqueness and measurability of solutions under some regularity assumptions and provide some estimates for the solutions. We moreover prove the stability with respect to the graphon particle systems and obtain the convergence of an interacting mean-field particle system with inhomogeneous interactions to the graphon mean-field BSDE. We then provide some comparison theorems for the graphon mean-field BSDEs. As an application, we introduce the graphon dynamic risk measure induced by the solution of a graphon mean-field BSDE system and study its properties. We finally provide a dual representation theorem for the graphon dynamic risk measure in the convex case.

]]>In this project, we consider a class of generalized Kyle-Back strategic insider trading models in which the insider is able to use the dynamic information obtained by observing the instantaneous movement of an underlying asset that is allowed to be influenced by its market price. Since such a model will be largely outside the Gaussian paradigm, we shall try to Markovize it by introducing an auxiliary (factor) diffusion process, in the spirit of the weighted total order process, as a part of the "pricing rule". As the main technical tool in solving the Kyle-Back equilibrium in such a setting, we study a class of Stochastic Two-Point Boundary Value Problem (STPBVP), which resembles the dynamic Markov bridge in the literature, but without insisting on its local martingale requirement. In the case when the solution of the STPBVP has an affine structure, we show that the pricing rule functions, whence the Kyle-Back equilibrium, can be determined by the decoupling field of a forward-backward SDE obtained via a non-linear filtering approach, along with a set of compatibility conditions. This is a joint work with Jin Ma.

]]>We will continue our study of continuous logic by

introducing model-theoretic notions such as theories and elementary

equivalence and the Tarski-Vaught test for continuous logic.

We discuss quantum ergodicity in the Benjamini-Schramm limit. This concerns equidistribution of eigenfunctions of Laplacian-like operators on sequences of spaces which ``converge'' to their common universal cover. We shall be particularly interested in the case when the universal cover is a symmetric space or an affine building (the non-archimedean analogue of a symmetric space). A result of this kind was first proven by Anantharaman-Le Masson for regular graphs and for which the underlying Laplacian-like operator is the adjacency operator. This result was reproven by Brooks-Le Masson-Lindenstrauss using a new technique which has been subsequently adapted to also work for rank one locally symmetric spaces (Le Masson-Sahlsten, Abert-Bergeron-Le Masson) and for higher rank locally symmetric spaces associated to $SL(d, R)$ (Brumley-Matz). We have obtained analogous results for Bruhat-Tits buildings associated to $SL(3, F)$ where $F$ is a non-archimedean local field. We shall discuss the strategy of proof common to all of these examples as well as discuss some of the new techniques introduced to handle the $SL(3, F)$ case.

]]>

First Step sessions are a great opportunity to learn more about the application process prior to meeting with an advisor. You can learn about all of our programs around the world, scholarships and other financial aid resources, the CGIS application process, and more!

*Attending a First Step session is no longer a required component of the CGIS application process.*

Define test configurations; Rees construction; Define the Futaki invariant; Intersection-theoretic formula for the Futaki invariant; Definition of K-stability; Examples.

]]>This is the second of a 2-3 part series on elementary Teichmüller theory, and on the SL(2, R) action on related moduli spaces. We will also explore links between the dynamics on Teichmüller space to the dynamics of discrete subgroups of Lie groups on homogeneous spaces.

]]>In their seminal work, Polyak and Juditsky showed that stochastic approximation algorithms for solving smooth equations enjoy a central limit theorem. Moreover, it has since been argued that the asymptotic covariance of the method is best possible among any estimation procedure in a local minimax sense of H´ajek and Le Cam. A long-standing open question in this line of work is whether similar guarantees hold for important non-smooth problems, such as stochastic nonlinear programming or stochastic variational inequalities. In this work, we show that this is indeed the case. This is joint work with Damek Davis and Liwei Jiang.

]]>Ruled surface is one of the most concrete examples we see when studying algebraic surfaces. These are surfaces that admit a fibration by $\PP^{1}$ over a curve. However, ruled surfaces are very boring since they are too easy in various ways. For instance, they can be easily classified, there are no degenerations, the algebraic structure of the fibre does not change, and the canonical bundle is easy to describe.

On the other hand, fibration by elliptic curves is way more entertaining since there are lots of things happening! We can study how the algebraic structure varies, how the fibre degenerates to a singular one, and can describe the canonical bundle in terms of the singular fibres and the moduli.

It turns out that these phenomena for elliptic surfaces can be generalized to many deep results in algebraic geometry such as variation of Hodge structure, degeneration of Hodge structure, adjunction and subadjunction, canonical bundle formula, semipositivity theorems, volume asymptotics and so on.

Despite the fact that elliptic fibrations are related to these profound theories in algebraic geometry, the example itself is very classical and can be understood explicitly. I will talk about these phenomena for elliptic surfaces in various perspectives.

First Step sessions are a great opportunity to learn more about the application process prior to meeting with an advisor. You can learn about all of our programs around the world, scholarships and other financial aid resources, the CGIS application process, and more!

*Attending a First Step session is no longer a required component of the CGIS application process.*

This talk focuses on q-orthogonal polynomials, orthogonal polynomials whose orthogonality condition is supported on the discrete lattice: q^k, for integer k . We investigate the behaviour of such polynomials in the case k>0 (q<1) and, if time permits, we study what happens when we relax this condition and allow k to be negative. Using the RHP framework we deduce the asymptotic behaviour of these polynomials as their degree tends to infinity, as well as other properties such as uniquness.

]]>TBA

]]>Title: Orbital integrals for gln and smoothening

Abstract: Orbital integral is a fundamental object in the geometric side of the trace formula. A traditional method to study orbital integrals is through Bruhat-Tits building or affine Springer fiber. In this talk, we will propose another method to study orbital integrals using smoothening.

As an application, we will explain a closed formula of the orbital integral for gln with n=2,3 and a new lower bound for a general n. We also propose a conjecture about the estimation of the orbital integral for any n. Our method works for any local field of characteristic 0 or >n. This is a joint work with Sungmun Cho.

The general subject of the talk is spectral theory of discrete (tight-binding) Schrodinger operators on $d$-dimensional lattices. For operators with periodic potentials, it is known that the spectra of such operators are purely absolutely continuous. For random i.i.d. potentials, such as the Anderson model, it is expected and can be proved in many cases that the spectra are almost surely purely point with exponentially decaying eigenfunctions (Anderson localization). Quasiperiodic operators can be placed somewhere in between: depending on the potential sampling function and the Diophantine properties of the frequency and the phase, one can have a large variety of spectral types. We will consider quasiperiodic operators

$$

(H(x)\psi)_n=\epsilon(\Delta\psi)_n+f(x+n\cdot\omega)\psi_n,\quad n\in \mathbb Z^d,

$$

where $\Delta$ is the discrete Laplacian, $\omega$ is a vector with rationally independent components, and $f$ is a $1$-periodic function on $\mathbb R$, monotone on $(0,1)$ with a positive lower bound on the derivative and some additional regularity properties. We will focus on two methods of proving Anderson localization for such operators: a perturbative method based on direct analysis of cancellations in the Rayleigh—Schr\”odinger perturbation series for arbitrary $d$, and a non—perturbative method based on the analysis of Green’s functions for $d=1$, originally developed by S. Jitomirskaya for the almost Mathieu operator.

The talk is based on joint works with S. Krymskii, L. Parnovski, and R. Shterenberg (perturbative methods) and S. Jitomirskaya (non-perturbative methods).

TBA

]]>Abstract: Two salient features of empirical temporal (i.e., time-varying) network data are the time-varying nature of network structure itself and heavy-tailed distributions of inter-contact times. Both of them can strongly impact dynamical processes occurring on networks, such as contagion processes, synchronization dynamics, and random walks. In the first part of the talk, I introduce theoretical explanation of heavy-tailed distributions of inter-contact times by state-dynamics modeling approaches in which each node is assumed to switch among a small number of discrete states in a Markovian manner and the nodes' states determine time-dependent edges. This approach is interpretable, facilitates mathematical analyses, and seeds various related mathematical modeling, algorithms, and data analysis (e.g., theorizing on epidemic thresholds, random walks on metapopulation models, inference of mixtures of exponential distributions, new Gillespie algorithms, embedding of temporal network data), some of which we will also discuss. The second part of the talk is on modeling of temporal networks by static networks that switch from one to another at regular time intervals. This approach facilitates analytical understanding of diffusive and epidemic dynamics on temporal networks as well as an efficient algorithm for containing epidemic spreading as convex optimization. Finally, I will touch upon some of my interdisciplinary collaborations including those on static networks.

Event will take place in-person in 4448 East Hall and online via Zoom.

Zoom Webinar Link:

https://umich.zoom.us/j/98734707290

Abstract:

Most disease associated genomic variants have relatively modest effects on target gene expression in reporter or CRISPR perturbation assays. In addition, enhancer disruption in vivo often has surprisingly weak phenotypic consequences. I will present machine learning (ML) methods (gkm-SVM and DNN) which we use to learn the complex transcription factor combinations that control enhancer activity and cell fate. I will then use these methods to develop a quantitative model for enhancer activity which shows that while promoter knockdown has robust effects on target gene expression, individual enhancer knockdown is often weaker and affects temporal transition dynamics, but not the final steady state. This model provides an explanation of the paradox of how enhancer variation can be strongly associated with disease risk while having individually weak effects, by showing in detail how gene regulatory networks control developmentally important and disease relevant cell state transitions and cancer.

https://umich-health.zoom.us/j/93929606089?pwd=SHh6R1FOQm8xMThRemdxTEFMWWpVdz09

First Step sessions are a great opportunity to learn more about the application process prior to meeting with an advisor. You can learn about all of our programs around the world, scholarships and other financial aid resources, the CGIS application process, and more!

*Attending a First Step session is no longer a required component of the CGIS application process.*

Introduce the space of valuations; Central fiber of test configurations viewed as divisorial valuations.

]]>TBA

]]>TBD

]]>

First Step sessions are a great opportunity to learn more about the application process prior to meeting with an advisor. You can learn about all of our programs around the world, scholarships and other financial aid resources, the CGIS application process, and more!

*Attending a First Step session is no longer a required component of the CGIS application process.*

TBA

]]>Title: TBA

Abstract: TBA

TBA

]]>MPSDS JPSM Seminar Series

February 15, 2023

12:00 - 1:00 EST

Richard Valliant, PhD, is a research professor emeritus at the Institute for Social Research, University of Michigan, and at the Joint Program in Survey Methodology at the University of Maryland. He is a Fellow of the American Statistical Association, an elected member of the International Statistical Institute, and has been an associate editor of the Journal of the American Statistical Association, Journal of Official Statistics, and Survey Methodology.

The Evolution of the Use of Models in Survey Sampling

The use of models in survey estimation has evolved over the last five (or more) decades. This talk will trace some of the developments over time and attempt to review some of the history. Consideration of models for estimating descriptive statistics began as early as the 1940's when Cochran and Jessen proposed linear regression estimators of means. These were early examples of model-assisted estimation since the properties of the Cochran-Jessen estimators were calculated with respect to a random sampling distribution. Model-thinking was used informally through the 1960's to form ratio and linear regression estimators that could in some applications reduce design variances.

In a 1963 Australian Journal of Statistics paper, Brewer presented results for a ratio estimator that were entirely based on a super population model. Royall (Biometrika 1970 and later papers) formalized the theory for a more general prediction approach using linear models. Since that time, the use of models is ubiquitous in the survey estimation literature and has been extended to nonparametric, empirical likelihood, Bayesian, small area, machine learning, and other approaches. There remains a considerable gap between the more advanced techniques in the literature and the methods commonly used in practice.

In parallel to the model developments, the design-based, randomization approach was dominating official statistics in the US largely due to the efforts of Morris Hansen and his colleagues at the US Census Bureau. In 1937 Hansen and others at the Census Bureau designed a follow-on sample survey to a special census of the employed and partially employed because response to the census was incomplete and felt to be inaccurate. The sample estimates were judged to be more trustworthy than those of the census itself. This began Hansen’s career-long devotion to random sampling as the only trustworthy method for obtaining samples from finite populations and for making inferences.

Model-assisted estimation, as discussed in the 1992 book by Särndal, Swensson, and Wretman is a type of compromise where models are used to construct estimators while a randomization distribution is used to compute properties like means and variances. This thinking has led to the popularity of doubly robust approaches where the goal is to have estimators with good properties with respect to both a randomization and a model distribution.

The field has now reached a troubling crossroads in which response rates to many types of surveys have plummeted and nonprobability datasets are touted as a way of obtaining reasonable quality data at low cost. Sophisticated model-based mathematical methods have been developed for estimation from nonprobability samples. In some applications, e.g., administrative data files that are incomplete due to late reporting, these methods may work well. However, in others the quality of nonprobability sample data is irremediably bad as illustrated by Kennedy in her 2022 Hansen lecture. In some situations, we are back in Morris' 1937 situation where standard approaches no longer work. Methods are needed to evaluate whether acceptable estimates can be made from the most suspect data sets. Nonetheless. nonprobability datasets are readily available now, and it is up to the statistical profession to develop good methods for using them.

Michigan Program in Survey and Data Science (MPSDS)

The University of Michigan Program in Survey Methodology was established in 2001 seeking to train future generations of survey and data scientists. In 2021, we changed our name to the Michigan Program in Survey and Data Science. Our curriculum is concerned with a broad set of data sources including survey data, but also including social media posts, sensor data, and administrative records, as well as analytic methods for working with these new data sources. And we bring to data science a focus on data quality — which is not at the center of traditional data science. The new name speaks to what we teach and work on at the intersection of social research and data. The program offers doctorate and master of science degrees and a certificate through the University of Michigan. The program's home is the Institute for Social Research, the world's largest academically-based social science research institute.

Summer Institute in Survey Research Techniques (SISRT)

The mission of the Summer Institute is to provide rigorous and high quality graduate training in all phases of survey research. The program teaches state-of-the-art practice and theory in the design, implementation, and analysis of surveys. The Summer Institute in Survey Research Techniques has presented courses on the sample survey since the summer of 1948, and has offered such courses every summer since. Graduate-level courses through the Program in Survey and Data Science are offered from June 5 through July 28 and available to enroll in as a Summer Scholar.

TBA

]]>TBA

]]>

First Step sessions are a great opportunity to learn more about the application process prior to meeting with an advisor. You can learn about all of our programs around the world, scholarships and other financial aid resources, the CGIS application process, and more!

*Attending a First Step session is no longer a required component of the CGIS application process.*

S (volume) and T-invariants of a filtration; Define uniform K-stability.

]]>TBA

]]>TBA

]]>TBD

]]>N/A

]]>

First Step sessions are a great opportunity to learn more about the application process prior to meeting with an advisor. You can learn about all of our programs around the world, scholarships and other financial aid resources, the CGIS application process, and more!

*Attending a First Step session is no longer a required component of the CGIS application process.*

TBA

]]>Title: TBA

Abstract: TBA

TBA

]]>

First Step sessions are a great opportunity to learn more about the application process prior to meeting with an advisor. You can learn about all of our programs around the world, scholarships and other financial aid resources, the CGIS application process, and more!

*Attending a First Step session is no longer a required component of the CGIS application process.*

K-stability of canonically polarized and Calabi-Yau varieties; A necessary condition for K-stability.

]]>Topic: *Modeling CTL-mediated Tumor Cell Death Mechanisms and the Activity of Immune Checkpoints in Immunotherapy*

Immunotherapy has dramatically transformed the cancer treatment landscape. Of the variety of types of immunotherapies available, immune checkpoint inhibitors (ICIs), which block inhibitory signals from tumor cells and reinvigorate killing activities of immune cells, have gained the spotlight. Although ICIs have shown promising results for some patients, the low response rates in many cancers highlight the challenges of using immune checkpoint blockade as an effective treatment. Cytotoxic T lymphocytes (CTLs) execute their cell-killing function via two distinct mechanisms. The first process is fast-acting and perforin/granzyme-mediated, and the second is a slower, Fas ligand (FasL)-driven killing mechanism. There is also evidence suggesting that the preferred killing mechanism by CTLs depends on the antigenicity of tumor cells. To determine the key factors affecting responses to checkpoint blockade therapy, we constructed an ordinary differential equation model describing in vivo tumor-immune dynamics in the presence of active or blocked PD-1/PDL1 immune checkpoint. Specifically, we analyzed which aspects of the tumor-immune landscape affect the response to ICIs with endpoints of tumor size and composition in the short and long term. By generating a virtual cohort with heterogeneous tumor and immune attributes, we also simulated the therapeutic outcomes of immune checkpoint blockade in a largely diverse population. In this way, we identified key tumor and immune characteristics that are associated with tumor elimination, dormancy and escape. This talk will also shed light on which fraction of a population potentially responds well to ICIs and ways to enhance therapeutic outcomes with combination therapy.

We develop different models to study the flutter of membranes (of zero bending rigidity) with vortex-sheet wakes in two- and three-dimensional inviscid flows. For 2D flows, we use a nonlinear, time-stepping method to study large-amplitude dynamics in the space of three dimensionless parameters: membrane pretension, mass density, and stretching rigidity. With a linearized version of the membrane-vortex-sheet model we also investigate the instability of a membrane by solving a nonlinear eigenvalue problem iteratively, for three boundary conditions---both ends fixed, one end fixed and one free, and both free. We further consider a simple physical setup: a membrane held by tethers with hinged ends, that interpolates between the fixed--fixed and free--free cases. We additionally study an infinite membrane model mounted on a periodic array of Hookean springs. This model allows us to compute asymptotic scaling laws for how the frequencies, growth rates, and eigenmodes depend on membrane pretension and mass density. Finally, we develop a nonlinear model and computational method to study large-amplitude membrane flutter in 3D inviscid flow for 12 different boundary conditions.

]]>TBD

]]>N/A

]]>

First Step sessions are a great opportunity to learn more about the application process prior to meeting with an advisor. You can learn about all of our programs around the world, scholarships and other financial aid resources, the CGIS application process, and more!

*Attending a First Step session is no longer a required component of the CGIS application process.*

TBA

]]>MPSDS JPSM Seminar Series

March 8, 2023

12:00 - 1:00 EST

Abstract

Respondent driven sampling (RDS) is a sampling method that leverages the respondents' networks to reach more members of the target population. In RDS, the size of the respondents' social network (also known as personal network size (PNS), or respondent's degree) is important in both the study operations and in estimation. A commonly used estimation of degree is the self-reported data from the interview, which typically has substantial measurement error, and, specifically, is found to be frequently rounded to a multiple of five. Measurement error in the PNS can introduce biased estimates for RDS, especially if the misreporting of the degree is associated with the outcome to be estimated.

This brown bag will present two related studies on the measurement of PNS. The first study uses two sets of data; 1) semi-structured in-depth interviews conducted over Zoom with 19 adult respondents of various ages, gender identities (transgender, nonbinary, cisgender), race, and sexual orientations (gay, lesbian, bi), 2) an RDS web survey targeting the adult LGBT population (n = 394). Thematic analysis conducted on the semi-structured interview transcripts showed a large variation in how respondents define "knowing" someone; for some respondents, it covers a larger network than the "recruitable" network (the network of people respondents are likely to think of recruiting to an RDS study). Meanwhile, the web-RDS shows that the more restrictive PNS questions yielded more realistic ranges for a "recruitable" network, with less proportion of rounded responses on the more restrictive PNS questions.

Motivated by the desire to improve the degree estimation in RDS, the second study presents a latent variable model to make inferences about participants’ actual degrees and potential reporting behaviors. Specifically, individual-level degree estimation will be obtained by revealing the association between the actual degree and relevant personal characteristics and blending their response to “How many [a particular sub-population] do you know in the target population?” Simulation studies demonstrate that the proposed method delivers sensible estimations about the individual degree.

Bios

Ai Rene Ong works at American Institutes for Research (AIR) as a Researcher/Survey Methodologist in the area of Education Statistics. She graduated with a PhD in Survey Methodology from the University of Michigan in 2022. Her dissertation research was on the measurement of network size and the mechanism of peer recruitment in Respondent Driven Sampling — a sampling method typically used for hard-to-sample populations.

Yibo Wang is a 3rd year Ph.D. candidate from the department of Biostatistics. She is now working with Dr. Sunghee Lee and Dr. Michael Elliott on measurement estimation in Respondent Driven Sampling

Michigan Program in Survey and Data Science (MPSDS)

The University of Michigan Program in Survey Methodology was established in 2001 seeking to train future generations of survey and data scientists. In 2021, we changed our name to the Michigan Program in Survey and Data Science. Our curriculum is concerned with a broad set of data sources including survey data, but also including social media posts, sensor data, and administrative records, as well as analytic methods for working with these new data sources. And we bring to data science a focus on data quality — which is not at the center of traditional data science. The new name speaks to what we teach and work on at the intersection of social research and data. The program offers doctorate and master of science degrees and a certificate through the University of Michigan. The program's home is the Institute for Social Research, the world's largest academically-based social science research institute.

Summer Institute in Survey Research Techniques (SISRT)

The mission of the Summer Institute is to provide rigorous and high quality graduate training in all phases of survey research. The program teaches state-of-the-art practice and theory in the design, implementation, and analysis of surveys. The Summer Institute in Survey Research Techniques has presented courses on the sample survey since the summer of 1948, and has offered such courses every summer since. Graduate-level courses through the Program in Survey and Data Science are offered from June 5 through July 28 and available to enroll in as a Summer Scholar.

The Summer Institute uses the sample survey as the basic instrument for the scientific measurement of human activity. It presents sample survey methods in courses designed to meet the educational needs of those specializing in social and behavioral research such as professionals in business, public health, natural resources, law, medicine, nursing, social work, and many other domains of study.

First Step sessions are a great opportunity to learn more about the application process prior to meeting with an advisor. You can learn about all of our programs around the world, scholarships and other financial aid resources, the CGIS application process, and more!

*Attending a First Step session is no longer a required component of the CGIS application process.*

Special test configurations; $\beta$-invariant and Fujita-Li criterion; $\alpha$ and $\delta$-invariants.

]]>N/A

]]>

First Step sessions are a great opportunity to learn more about the application process prior to meeting with an advisor. You can learn about all of our programs around the world, scholarships and other financial aid resources, the CGIS application process, and more!

*Attending a First Step session is no longer a required component of the CGIS application process.*

TBA

]]>Title: TBA

Abstract: TBA

TBA

]]>MPSDS JPSM Seminar Series

March 15, 2023

12:00 - 1:00 EST

Annette Jäckle is Professor of Survey Methodology at the Institute for Social and Economic Research at the University of Essex, UK and Associate Director of Innovations and Co-Investigator of the UK Household Longitudinal Study: Understanding Society. Her research interests are in methodology of data collection for longitudinal studies, mixed mode data collection, questionnaire design, respondent consent to data linkage, and new ways of using mobile devices for survey data collection.

Abstract

Data linkage usually requires informed consent of respondents, whether for legal or ethical reasons. A common problem is that when consent questions are asked in self-completion surveys, respondents are much less likely to consent than when they are asked for consent in interviewer administered surveys. In the existing literature, predictors of consent are mostly inconsistent, between studies, but also between different consents asked within one study. In addition, experiments with the wording of consent questions have often had no or inconsistent effects. Why is this? And what can be done to increase informed consent to data linkage? This presentation provides an overview of what we have learnt from qualitative in-depth interviews and a series of experiments implemented in two UK probability household panels (the Understanding Society Innovation Panel and COVID-19 study) and in the UK PopulusLive online access panel. We address the following questions. (1) How do respondents decide whether to consent to data linkage? (2) Why are respondents less likely to consent in web than CAPI surveys? (3) How best to ask for multiple consents within a survey? (4) Which wording and formats affect informed consent and why? We end the overview with a summary of the practical implications for how best to ask for consent to data linkage.

Michigan Program in Survey and Data Science (MPSDS)

The University of Michigan Program in Survey Methodology was established in 2001 seeking to train future generations of survey and data scientists. In 2021, we changed our name to the Michigan Program in Survey and Data Science. Our curriculum is concerned with a broad set of data sources including survey data, but also including social media posts, sensor data, and administrative records, as well as analytic methods for working with these new data sources. And we bring to data science a focus on data quality — which is not at the center of traditional data science. The new name speaks to what we teach and work on at the intersection of social research and data. The program offers doctorate and master of science degrees and a certificate through the University of Michigan. The program's home is the Institute for Social Research, the world's largest academically-based social science research institute.

Summer Institute in Survey Research Techniques (SISRT)

The mission of the Summer Institute is to provide rigorous and high quality graduate training in all phases of survey research. The program teaches state-of-the-art practice and theory in the design, implementation, and analysis of surveys. The Summer Institute in Survey Research Techniques has presented courses on the sample survey since the summer of 1948, and has offered such courses every summer since. Graduate-level courses through the Program in Survey and Data Science are offered from June 5 through July 28 and available to enroll in as a Summer Scholar.

The Summer Institute uses the sample survey as the basic instrument for the scientific measurement of human activity. It presents sample survey methods in courses designed to meet the educational needs of those specializing in social and behavioral research such as professionals in business, public health, natural resources, law, medicine, nursing, social work, and many other domains of study.

Given two probability measures on sequential data, we investigate the transport problem with time-inconsistent preferences under a discrete-time setting. Motivating examples include nonlinear objectives, state-dependent costs, and regularized optimal transport with general $f$-divergence. Under the bi-causal constraint, we introduce equilibrium transport and characterize it with maximum theorem and extended dynamic programming principle. We apply our framework to study the state dependence of two job markets including top-ranking executives and academia. The empirical analysis shows that a job market with a stronger state dependence is less efficient. The University of California (UC) postdoc job market has the strongest state dependence even than that of top executives, while there is no evidence of state dependence on the UC faculty job market. This is a joint work with Erhan Bayraktar.

]]>

First Step sessions are a great opportunity to learn more about the application process prior to meeting with an advisor. You can learn about all of our programs around the world, scholarships and other financial aid resources, the CGIS application process, and more!

*Attending a First Step session is no longer a required component of the CGIS application process.*

TBA

]]>Variational approach; Valuations computing $\delta$-invariant; Finite generation.

]]>TBA

]]>TBD

]]>N/A

]]>

First Step sessions are a great opportunity to learn more about the application process prior to meeting with an advisor. You can learn about all of our programs around the world, scholarships and other financial aid resources, the CGIS application process, and more!

*Attending a First Step session is no longer a required component of the CGIS application process.*

TBA

]]>Title: TBA

Abstract: TBA

TBA

]]>

First Step sessions are a great opportunity to learn more about the application process prior to meeting with an advisor. You can learn about all of our programs around the world, scholarships and other financial aid resources, the CGIS application process, and more!

*Attending a First Step session is no longer a required component of the CGIS application process.*

TBA

]]>Overview; Normalized volume (existence and uniqueness of the minimizer); Boundedness via normalized volume.

]]>TBA

]]>TBA

]]>TBD

]]>N/A

]]>

First Step sessions are a great opportunity to learn more about the application process prior to meeting with an advisor. You can learn about all of our programs around the world, scholarships and other financial aid resources, the CGIS application process, and more!

*Attending a First Step session is no longer a required component of the CGIS application process.*

TBA

]]>Title: TBA

Abstract: TBA

TBA

]]>TBA

]]>I will talk about the following question of Gromov: what closed manifolds can be efficiently wrapped with Euclidean wrapping paper? That is, for what M is there a 1-Lipschitz map $\mathbb R^n \to M$ with positive asymptotic degree? Gromov called such manifolds elliptic. We show that, for example, the connected sum of k copies of CP^2 is elliptic if and only if k ≤ 3. I will try to explain the intuition behind this example, how it extends to a more general dichotomy governed by the de Rham cohomology of M, and why ellipticity is central to the program of understanding the relationship between topology and metric properties of maps.

If I have time, I'll also explain why for a non-elliptic M, a maximally efficient map $\mathbb R^n \to M$ must have components at many different frequencies (in a Fourier-analytic sense), and even then it's at best logarithmically far from having positive asymptotic degree. This is joint work with Sasha Berdnikov and Larry Guth.

First Step sessions are a great opportunity to learn more about the application process prior to meeting with an advisor. You can learn about all of our programs around the world, scholarships and other financial aid resources, the CGIS application process, and more!

*Attending a First Step session is no longer a required component of the CGIS application process.*

TBA

]]>Openness (via normalized volume); existence of a good moduli; properness and projectivity of K-moduli.

]]>TBA

]]>TBD

]]>Note unusual time.

]]>N/A

]]>

First Step sessions are a great opportunity to learn more about the application process prior to meeting with an advisor. You can learn about all of our programs around the world, scholarships and other financial aid resources, the CGIS application process, and more!

*Attending a First Step session is no longer a required component of the CGIS application process.*

Title: TBA

Abstract: TBA

MPSDS JPSM Seminar Series

April 5, 2022

12:00 - 1:00 EST

Stephanie Morales is a second-year Ph.D. student at the University of Michigan's Program in Survey and Data Science. She holds a BA in Psychology and an MA in Sociology. She is interested in cross-cultural surveys, measurement error in data collection with racial/ethnic minorities, and adaptive survey design.

Assessing Cross-Cultural Comparability of Self-Rated Health and Its Conceptualization through Web Probing

Self-rated health (SRH) is a widely used question across different fields, as it is simple to administer yet has been shown to predict mortality. SRH asks respondents to rate their overall health typically using Likert-type response scales (i.e., excellent, very good, good, fair, poor). Although SRH is commonly used, few studies have examined its conceptualization from the respondents’ point of view and even less so for differences in its conceptualization across diverse populations. We aim to assess the comparability of SRH across different cultural groups by investigating the factors that respondents consider when responding to the SRH question. We included an open-ended probe asking what respondents thought when responding to SRH in web surveys conducted in five countries: Great Britain, Germany, the U.S., Spain, and Mexico. In the U.S., we targeted six racial/ethnic and linguistic groups: English-dominant Koreans, Korean-dominant Koreans, English-dominant Latinos, Spanish-dominant Latinos, non-Latino Black Americans, and non-Latino White Americans. One novelty of our study is allowing multiple attribute codes (e.g., health behaviors, illness) per respondent and tone (e.g., in the direction of positive or negative health or neutral) of the probing responses for each attribute, allowing us 1) to assess respondents’ thinking process holistically and 2) to examine whether and how respondents mix attributes. Our study compares the number of reported attributes and tone by cultural groups and integrates SRH responses in the analysis. This study aims to provide a deeper understanding of SRH by revealing the cognitive processes among diverse populations and is expected to shed light on its cross-cultural comparability.

Michigan Program in Survey and Data Science (MPSDS)

The University of Michigan Program in Survey Methodology was established in 2001 seeking to train future generations of survey and data scientists. In 2021, we changed our name to the Michigan Program in Survey and Data Science. Our curriculum is concerned with a broad set of data sources including survey data, but also including social media posts, sensor data, and administrative records, as well as analytic methods for working with these new data sources. And we bring to data science a focus on data quality — which is not at the center of traditional data science. The new name speaks to what we teach and work on at the intersection of social research and data. The program offers doctorate and master of science degrees and a certificate through the University of Michigan. The program's home is the Institute for Social Research, the world's largest academically-based social science research institute.

Summer Institute in Survey Research Techniques (SISRT)

The mission of the Summer Institute is to provide rigorous and high quality graduate training in all phases of survey research. The program teaches state-of-the-art practice and theory in the design, implementation, and analysis of surveys. The Summer Institute in Survey Research Techniques has presented courses on the sample survey since the summer of 1948, and has offered such courses every summer since. Graduate-level courses through the Program in Survey and Data Science are offered from June 5 through July 28 and available to enroll in as a Summer Scholar.

The Summer Institute uses the sample survey as the basic instrument for the scientific measurement of human activity. It presents sample survey methods in courses designed to meet the educational needs of those specializing in social and behavioral research such as professionals in business, public health, natural resources, law, medicine, nursing, social work, and many other domains of study.

Abstract: TBA

]]>

First Step sessions are a great opportunity to learn more about the application process prior to meeting with an advisor. You can learn about all of our programs around the world, scholarships and other financial aid resources, the CGIS application process, and more!

*Attending a First Step session is no longer a required component of the CGIS application process.*

Speaker: Marius Beceanu (SUNY, Albany)

]]>TBD

]]>TBA

]]>TBD

]]>N/A

]]>

First Step sessions are a great opportunity to learn more about the application process prior to meeting with an advisor. You can learn about all of our programs around the world, scholarships and other financial aid resources, the CGIS application process, and more!

*Attending a First Step session is no longer a required component of the CGIS application process.*

MPSDS JPSM Seminar Series

April 12, 2023

12:00 - 1:00 EST

Michael Elliott is a Professor of Biostatistics and a Research Professor at the Institute for Social Research. Dr. Elliott's research interests include the design and analysis of sample surveys, causal inference, and missing and latent variable data structures.

Abstract

Medical researchers have understood for many years that treatment effect estimates obtained from a randomized clinical trial (RCT) -- termed efficacy'' -- can differ from those obtained in a general population -- termed effectiveness''. Only in the past decade has extensive work begun in the statistical literature to bridge this gap using formal quantitative methods. As noted by Rod Little in a letter to the editor in the New Yorker ...randomization in randomized clinical trials concerns the allocation of the treatment, not the selection of individuals for the study. The latter can have an important impact on the average size of a treatment effect,'' with RCT samples often designed, sometimes explicitly, to be more likely to include individuals for whom the treatment may be more effective.

This issue has been various termed generalizability'' or transportability." Why do we care about transportability? In RCTs we are in the happy situation were treatment assignment is randomized, so confounding due to either observed or unobserved (pre-treatment) covariates is not an issue. But while randomization of treatment eliminates the effect of unobserved confounders, at least net of non-compliance, it does not eliminate the effect of unobserved effect modifiers, which can impact the causal effect of treatment in a population that differs from the RCT sample population. The impact of these interactions on the marginal effect of treatment thus can differ between the RCT population and the final population of interest.

Concurrent with research into transportability has been research into making population inference from non-probability samples. There is a close overlap between these two approaches, particularly with respect to the non-probability inference methods that rely on information from a relevant probability sample of the target population to reduce selection bias effects. When there are relevant censuses or probability samples of the target patient population of interest, these methods can be adapted to transport information from the RCT to the patient population. Because the RCT setting focuses on causal inference, this adaptation involves extensions to estimate counterfactuals. Thus approaches that treat population inference as a missing data problem are a natural fit to connect these two strands of methodological innovation.

In particular, we propose to extend a pseudo-weighting'' methodology from other non-probability settings to a doubly robust'' estimator that treats sampling probabilities or weights as regression covariates to achieve consistent estimation of population quantities. We explore our proposed approach and compare with some standard existing methods in a simulation study to assess the effectiveness of the approach under differing degrees of selection bias and model misspecification, and compare it with results obtained using the RT data only and with existing methods that use inverse probability weights. We apply it to a study of pulmonary artery catheterization in critically ill patients where we believe differences between the trial sample and the larger population might impact overall estimates of treatment effects.

MPSDS

The University of Michigan Program in Survey Methodology was established in 2001 seeking to train future generations of survey and data scientists. In 2021, we changed our name to the Michigan Program in Survey and Data Science. Our curriculum is concerned with a broad set of data sources including survey data, but also including social media posts, sensor data, and administrative records, as well as analytic methods for working with these new data sources. And we bring to data science a focus on data quality — which is not at the center of traditional data science. The new name speaks to what we teach and work on at the intersection of social research and data. The program offers doctorate and master of science degrees and a certificate through the University of Michigan. The program's home is the Institute for Social Research, the world's largest academically-based social science research institute.

Summer Institute in Survey Research Techniques (SISRT)

The mission of the Summer Institute is to provide rigorous and high quality graduate training in all phases of survey research. The program teaches state-of-the-art practice and theory in the design, implementation, and analysis of surveys. The Summer Institute in Survey Research Techniques has presented courses on the sample survey since the summer of 1948, and has offered such courses every summer since. Graduate-level courses through the Program in Survey and Data Science are offered from June 5 through July 28 and available to enroll in as a Summer Scholar.

The Summer Institute uses the sample survey as the basic instrument for the scientific measurement of human activity. It presents sample survey methods in courses designed to meet the educational needs of those specializing in social and behavioral research such as professionals in business, public health, natural resources, law, medicine, nursing, social work, and many other domains of study.

Abstract: TBA

]]>Thinking of studying abroad during the winter term? Visit the CGIS Office for our Winter Advising event to learn more about major-specific programs such as programs in the environment, Spanish, and Humanities/Social Sciences, and interest-specific program sessions such as studying abroad in the UK and English-taught programs in Asia to name a few. The event offers students the opportunity to interact with advisors and address questions or concerns you may have regarding studying abroad.

]]>

First Step sessions are a great opportunity to learn more about the application process prior to meeting with an advisor. You can learn about all of our programs around the world, scholarships and other financial aid resources, the CGIS application process, and more!

*Attending a First Step session is no longer a required component of the CGIS application process.*

TBA

]]>TBA

]]>N/A

]]>

First Step sessions are a great opportunity to learn more about the application process prior to meeting with an advisor. You can learn about all of our programs around the world, scholarships and other financial aid resources, the CGIS application process, and more!

*Attending a First Step session is no longer a required component of the CGIS application process.*

Title: TBA

Abstract: TBA