Time: 10/28/2016, 1:00-4:15pm
Place: W3008 (location changed)
Buffet lunch starts at 1:00pm.
Title: Functional and very high dimension reduction
Speaker: Yanyuan Ma, Department of Statistics, Pennsylvania State University
The talk has two components. In the first component,
to study the relation between a univariate response and multiple functional covariates, we propose a functional single index model that is semiparametric. The parametric part of the model integrates the linear regression modeling for functional data and the sucient
dimension reduction structure. The nonparametric part of the model further allows the response-index dependence or the link function to be unspecied. We use B-splines to approximate the coecient function in the functional linear regression model part and reduce the problem to a familiar dimension folding model. We develop a new method to handle the subsequent dimension folding model by using kernel regression in combination with semiparametric treatment. The new method does not impose any special requirement on the inner product between the covariate function and the B-spline bases, and allows ecient estimation of both the index vector and the B-spline coecients. The estimation method is general and applicable to both continuous and discrete response variables. We further derive asymptotic properties of the class of methods for both the index vector and the coecient function. We establish the semiparametric optimality, which has not been done before in a semiparametric model where both kernel and B-spline estimation are involved.
In the second component, we study large genetic data available easily due to technology advance. However, in comparison with the data collection procedure,
statistical analysis is still much cheaper. Thus, secondary analysis of SNPs data re-analyze existing data in an effort to extract more information, is attractive and cost effective. We study the relation between gene expression and SNPs through a combination of factor analysis and dimension reduction estimation (FADRE). To take advantage of the flexibility in traditional factor models where the latent factors are not required to be normal, we recommend using semiparametric sufficient dimension reduction methods in the joint estimation of the combined model. The resulting estimator is flexible and has superior performance. We further quantify the asymptotic performance of the parameter estimation and perform inference. The new results enables us to identify statistically significant SNPs concerning gene-SNPs relation in lung tissues for the first time
from GTEx data.
Joint SMART/SLAM Seminar
Title: Optimizing the Personalized Timing for Treatment Initiation with Random Decision Points
Speaker: Lu Wang, Department of Biostatistics, University of Michigan
Stepwise intensification of treatment is often necessary for chronic diseases with progressive conditions. An important but challenging problem is to find the optimal personalized timing to initiate a treatment for the next stage of disease condition. In this talk, we consider estimating the optimal dynamic treatment regimes (DTRs) to determine a personalized timing for treatment initiation given a patient’s specific characteristics. We aim to identify the optimal DTR amongst a set of regimes predefined by key biomarkers indicating disease severity, which are monitored continuously during a follow-up period. Instead of considering multiple fixed decision stages as in most DTR literature, our study undertakes the task of dealing with continuous random decision points for treatment initiation based on patients’ biomarker and treatment history. Under each candidate DTR, we employ a flexible survival model with splines of time-varying covariates to estimate the patient-specific probability of adherence to the regime. With the estimated probability, we construct an inverse probability weighted estimator for the counterfactual mean utility to assess the DTR. We conduct simulations to demonstrate the performance of our method and further illustrate the application process with an example of insulin therapy initiation among type 2 diabetic patients.
10/28/2016 3:00-4:15pm, Joint SMART/SLAM Seminar, Lu Wang, Department of Biostatistics, University of Michigan
11/4/2016 Jon Steingrimsson, Department of Biostatistics, JHU
11/11/2016 Jiawei Bai, Department of Biostatistics, JHU
11/18/2016 Yifei Sun, Department of Biostatistics, JHU
12/2/2016 Zonghui Hu, Biostatistics Research Branch, NIAID, NIH
2/3/2017 Lu Mao, University of Wisconsin
2/17/2017 Xuelin Huang, MD Anderson Cancer Center
3/24/2016 Judy Li, FDA