Events

A Tale of Two Parasites: how can Gaussian processes contribute to improved public health in Africa?

In this talk, I will rst make some general comments about the role of statistical modelling in scientic research, illustrated by two examples from infectious disease epidemiology. I will then describe in detail how statistical modelling based on Gaussian spatial stochastic processes has been used to construct region-wide risk maps to inform the operation of a multi-national control programme for onchocerciasis (river blindness) in equatorial Africa. Finally, I will describe work-in progress aimed at exploiting recent developments in mobile microscopy to enable more precise local predictions of community-level risk.

Challenges in Developing Learning Algorithms to Personalize Treatment in Real Time

A formidable challenge in designing sequential treatments is to  determine when and in which context it is best to deliver treatments.  Consider treatment for individuals struggling with chronic health conditions.  Operationally designing the sequential treatments involves the construction of decision rules that input current context of an individual and output a recommended treatment.   That is, the treatment is adapted to the individual's context; the context may include  current health status, current level of social support and current level of adherence for example.  Data sets on individuals with records of time-varying context and treatment delivery can be used to inform the construction of the decision rules.    There is much interest in personalizing the decision rules, particularly in real time as the individual experiences sequences of treatment.   Here we discuss our work in designing  online "bandit" learning algorithms for use in personalizing mobile health interventions. 

Assessing financial model risk

Model risk has a huge impact on any financial or insurance risk measurement procedure and its quantification is therefore a crucial step. In this talk, we introduce three quantitative measures of model risk when choosing a particular reference model within a given class: the absolute measure of model risk, the relative measure of model risk and the local measure of model risk. Each of the measures has a specific purpose and so allows for flexibility. We illustrate the various notions by studying some relevant examples, so as to emphasize the practicability and tractability of our approach.

Uncovering the Mechanisms of General Anesthesia: Where Neuroscience Meets Statistics

General anesthesia is a drug-induced, reversible condition involving unconsciousness, amnesia (loss of memory), analgesia (loss of pain sensation), akinesia (immobility), and hemodynamic stability. I will describe a primary mechanism through which anesthetics create these altered states of arousal. Our studies have allowed us to give a detailed characterization of the neurophysiology of loss and recovery of consciousness​, in the case of propofol, and we have demonstrated ​​ that the state of general anesthesia can be rapidly reversed by activating specific brain circuits. The success of our research has depended critically on tight coupling of experiments, ​statistical signal processing​​ and mathematical modeling.

A Machine Learning Approach to Portfolio Risk Management

Risk measurement, valuation and hedging form an integral task in portfolio risk management for insurance companies and other financial institutions. Portfolio risk arises because the values of constituent assets and liabilities change over time in response to changes in the underlying risk factors. The quantification of this risk requires modeling the dynamic portfolio value process. This boils down to compute conditional expectations of future cash flows over long time horizons, e.g., up to 40 years and beyond, which is computationally challenging. 

This lecture presents a framework for dynamic portfolio risk management in discrete time building on machine learning theory. We learn the replicating martingale of the portfolio from a finite sample of its terminal cumulative cash flow. The learned replicating martingale is in closed form thanks to a suitable choice of the reproducing kernel Hilbert space. We develop an asymptotic theory and prove
convergence and a central limit theorem. We also derive finite sample error bounds and concentration inequalities. As application we compute the value at risk and expected shortfall of the one-year loss of some stylized portfolios.

Optimal Transport, Entropy, and Risk Measures on Wiener space

We discuss the interplay between entropy, large deviations, and optimal couplings on Wiener space.

In particular we prove a new rescaled version of Talagrand’s transport inequality. As an application, we consider rescaled versions of the entropic risk measure which are sensitive to risks in the fine structure of Brownian paths. 

Precision Factor Investing: Avoiding Factor Traps by Predicting Heterogeneous Effects of Firm Characteristics

We apply ideas from causal inference and machine learning to estimate the sensitivity of future stock returns to observable characteristics like size, value, and momentum. By analogy with the informal notion of a "value trap," we distinguish "characteristic traps" (stocks with weak sensitivity) from "characteristic responders" (those with strong sensitivity). We classify stocks by interpreting these distinctions as heterogeneous treatment effects (HTE), with characteristics interpreted as treatments and future returns interpreted as responses. The classification exploits a large set of stock features and recent work applying machine learning to HTE. Long-short strategies based on sorting stocks on characteristics perform significantly better when applied to characteristic responders than traps. A strategy based on the difference between these long-short returns profits from the predictability of HTE rather than from factors associated with the characteristics themselves. This is joint work with Pu He.

Building Deep Statistical Thinking for Data Science 2020: Privacy Protected Census, Gerrymandering, and Election

The year 2020 will be a busy one for statisticians and more generally data scientists.  The US Census Bureau has announced that the data from the 2020 Census will be released under differential privacy (DP) protection, which in layperson’s terms means adding some noises to the data.  While few would argue against protecting data privacy, many researchers, especially from the social sciences, are concerned whether the right trade-offs between data privacy and data utility are being made. The DP protection also has direct impact on redistricting, an issue that is already complicated enough with accurate counts, due to the need of guarding against excessive gerrymandering.  The central statistical problem there is a rather unique one:  how to determine whether a realization is an outlier with respect to a null distribution, when that null distribution itself cannot be fully determined?  The 2020 US election will be another highly watched event, with many groups already busy making predictions. Will the lessons from predicting the 2016 US election be learned, or the failure be repeated?  This talk invites the audience on a journey of deep statistical thinking prompted by these questions, regardless whether they have any interest in the US Census or politics.

Navigation and Evaluation of Latent Structure in High-Dimensional Data

In the modern data analysis paradigm, fitting models is easy, but knowing how to design or evaluate them is difficult. In this talk, we will adapt insights from graphical statistics and goodness-of-fit testing to modern problems, illustrating them with applications to microbiome genomics and climate systems science.

For the microbiome, we show how linking complementary displays can make it easy to query structure in raw data. We also find novel visual summaries that inform model criticism more deeply than data splitting strategies alone. We then describe how artificial intelligence can be used to accelerate climate simulations, and introduce techniques for characterizing goodness-of-fit of the resulting models.

Viewed broadly, these projects provide opportunities for human interaction in the automated data processing regime, facilitating (1) streamlined navigation of data and (2) critical evaluation of models.

Renewable Estimation and Incremental Inference in Streaming Data Analysis

New data collection and storage technologies have given rise to a new field of streaming data analytics, including real-time statistical methodology for online data analyses. Streaming data refers to high-throughput recordings with large volumes of observations gathered sequentially and perpetually over time. Such type of data includes national disease registry, mobile health, and disease surveillance, among others. This talk primarily concerns the development of a fast real-time statistical estimation and inference method for regression analysis, with a particular objective of addressing challenges in streaming data storage and computational efficiency. Termed as renewable estimation, this method enjoys strong theoretical guarantees, including both asymptotic unbiasedness and estimation efficiency, and fast computational speed. The key technical novelty pertains to the fact that the proposed method uses current data and summary statistics of historical data. The proposed algorithm will be demonstrated in generalized linear models (GLM) for cross-sectional data. I will discuss both conceptual understanding and theoretical guarantees of the method and illustrate its performance via numerical examples. This is joint work with my supervisor Professor Peter Song.