Canada Spends Logo

Explore our Budget 2026 analysis

Grant: Discovery Grants Program - Individual

McGill University

$14,000.00

Statistical inference in finite mixture of regressions and mixture-of-experts models in high-dimensional spaces, and varying coefficient finite mixture of regression models

FY 2015-16

Summary
INTRODUCTION: <span style=""line-height: 1.5em;"">In recent years, we have witnessed the rise of large scale data, colloquially referred to as big data, in different fields of scientific research ranging from biology and medicine to engineering, the social sciences and econometrics. <span style=""line-height: 1.5em;"">A common statistical problem of interest in the analysis of such data is to model a response variable of interest as a function of a small subset of a large number of features<span style=""line-height: 1.5em;"">.<sub style=""line-height: 1.5em;""> <span style=""line-height: 1.5em;"">This<span style=""line-height: 1.5em;""> is referred to as a feature selection problem.<span style=""line-height: 1.5em;""> <span style=""line-height: 1.5em;"">In addition to noise accumulation and spurious correlation, unobserved heterogeneity in high-dimensional data makes the feature selection problem even harder. Finite mixture of regressions (FMR) and mixture-of-experts (MOE) are powerful statistical models for capturing heterogeneity in data. The first part of this proposal focuses on f<span style=""line-height: 1.5em;"">eature selection, estimation and post-selection inference problems in FMR/MOE. The second part concerns varying coefficient finite mixture of regression (VC-FMR) models in which regression coefficients change as smooth functions of an index variable such as time. For example, in market segmentation research, consumer preferences for products often change over time and across different market segments. VC-FMR models provide a natural tool for modeling such phenomena which involves heterogeneous functional data. However, m<span style=""line-height: 1.5em;"">ethodological and computational tools for these relatively new models are largely unexplored.  <span style=""line-height: 1.5em;""><strong style=""line-height: 1.5em;"">OBJECTIVES<span style=""line-height: 1.5em;"">: An emphasis of my research program is on developing sound statistical methodology and computationally efficient algorithms for estimation, feature selection, and also post-selection inference such as hypothesis testing and confidence intervals in FMR/MOE in high dimensions. Another focal point of my research concerns estimation and feature selection in VC-FMR. My longer term objectives focus on complex time series data and high-dimensional heterogeneous and dependent data.  METHODS: I will study the regularization techniques LASSO/SCAD for simultaneous parameter estimation and feature selection in FMR/MOE models in high dimensions. Coordinate descent-type expectation-maximization (EM) algorithms will be investigated for numerical computations. Post-selection inference such as hypothesis testing and confidence intervals for parameters in sparse FMR/MOE will be explored based on sample splitting techniques. Regularized local kernel likelihood-based methods will be used for <span style=""line-height: 1.5em;"">functional parameter estimation and <span style=""line-height: 1.5em;"">feature selection in VC-FMR models.  IMPACT: My<span style=""line-height: 1.5em;""> proposed research program will address unresolved statistical issues in FMR/MOE models in high dimensions as well as in VC-FMR models, and offer solutions to practical problems of interest to a broader statistical audience. The proposed methods could then immediately be used to solve scientific problems in areas such as biology, engineering, the health sciences, and marketing research.   <span style=""line-height: 1.5em;"">
Program
Discovery Grants Program - Individual
Location
Quebec, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data

As published

Application id
RGPIN-2015-03805
Name nom
Khalili, Abbas
Department département
Mathematics and Statistics - Mathematics and Statistics
Organization id
61
Institution établissement
McGill University
Province en
Québec
Province fr
Québec
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2015
Competition year année de concours
2015
Award amount
14000
Program id
RGPIN
Program name en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
Discovery grants programs
Group fr
Programmes de subventions à la découverte
Committee code
1508
Committee name en
Mathematics and Statistics
Committee name fr
Mathématiques et statistique
Area of application code
1205
Area of application group en
Advancement of knowledge
Area of application group fr
Avancement des connaissances
Area of application en
Mathematical sciences
Area of application fr
Sciences mathématiques
Research subject code
3009
Research subject group en
Statistics and probability
Research subject group fr
Statistique et probabilité
Research subject en
Parametric inference
Research subject fr
Inférence paramétrique
Partie
1
Nb partie
5
Application title
Statistical inference in finite mixture of regressions and mixture-of-experts models in high-dimensional spaces, and varying coefficient finite mixture of regression models
Keyword
NSERC
Application summary
<p><strong>INTRODUCTION</strong>: <span style=""line-height: 1.5em;"">In recent years, we have witnessed the rise of large scale data, colloquially referred to as <em>big data</em>, in different fields of scientific research ranging from biology and medicine to engineering, the social sciences and econometrics. </span><span style=""line-height: 1.5em;"">A common statistical problem of interest in the analysis of such data is to model a response variable of interest as a function of a <em>small subset</em> of a large number of features</span><span style=""line-height: 1.5em;"">.</span><sub style=""line-height: 1.5em;""> </sub><span style=""line-height: 1.5em;"">This</span><span style=""line-height: 1.5em;""> is referred to as a <em>feature selection problem</em>.</span><span style=""line-height: 1.5em;""> </span><span style=""line-height: 1.5em;"">In addition to noise accumulation and spurious correlation, <em>unobserved heterogeneity</em> in high-dimensional data makes the feature selection problem even harder. Finite mixture of regressions (FMR) and mixture-of-experts (MOE) are powerful statistical models for capturing heterogeneity in data. The first part of this proposal focuses on f</span><span style=""line-height: 1.5em;"">eature selection, estimation and post-selection inference problems in FMR/MOE. The second part concerns <em>varying coefficient</em> finite mixture of regression (VC-FMR) models in which regression coefficients change as smooth functions of an index variable such as time. For example, in market segmentation research, consumer preferences for products often change over time and across different market segments. VC-FMR models provide a natural tool for modeling such phenomena which involves <em>heterogeneous functional data</em>. However, m</span><span style=""line-height: 1.5em;"">ethodological and computational tools for these relatively new models are largely unexplored. </span></p> <p><span style=""line-height: 1.5em;""></span><strong style=""line-height: 1.5em;"">OBJECTIVES</strong><span style=""line-height: 1.5em;"">: An emphasis of my research program is on developing sound statistical methodology and computationally efficient algorithms for estimation, feature selection, and also post-selection inference such as hypothesis testing and confidence intervals in FMR/MOE in high dimensions. Another focal point of my research concerns estimation and feature selection in VC-FMR. My longer term objectives focus on complex time series data and high-dimensional <em>heterogeneous</em> and <em>dependent</em> data. </span></p> <p><strong>METHODS</strong>: I will study the regularization techniques LASSO/SCAD for simultaneous parameter estimation and feature selection in FMR/MOE models in high dimensions. Coordinate descent-type expectation-maximization (EM) algorithms will be investigated for numerical computations. Post-selection inference such as hypothesis testing and confidence intervals for parameters in sparse FMR/MOE will be explored based on sample splitting techniques. Regularized local kernel likelihood-based methods will be used for <span style=""line-height: 1.5em;"">functional parameter estimation and </span><span style=""line-height: 1.5em;"">feature selection in VC-FMR models. </span></p> <p> <strong>IMPACT</strong>: My<span style=""line-height: 1.5em;""> proposed research program will address unresolved statistical issues in FMR/MOE models in high dimensions as well as in VC-FMR models, and offer solutions to practical problems of interest to a broader statistical audience. The proposed methods could then immediately be used to solve scientific problems in areas such as biology, engineering, the health sciences, and marketing research.   </span><span style=""line-height: 1.5em;""> </span></p>