University of Pittsburgh Directory Calendar News Careers Alumni Giving

Search
  • Home
  • Behavioral and Community Health Sciences
  • Biostatistics
    • Academics

      • MS
      • MS - HDS concentration
      • MS - SCG concentration
      • PhD
      • Courses
        • Class Schedule
      • Student Handbooks & Forms
    • Research & Practice

      • Centers Overview
      • Faculty Research
      • Student Research
      • Consulting
      • Practice
    • Admissions & Aid

      • Admissions
      • Financial Aid
      • Frequently Asked Questions
      • Request Information
      • Apply Now
    • Seminars & Events

      • Biostatistics Events
      • Biostatistics Seminars
      • Biostatistics Dissertation Defenses
      • Conferences
        • ENAR Spring Meeting
        • JSM
    • Who's Who

      • Faculty (All)
        • Primary Faculty
        • Secondary Faculty
        • Adjunct Faculty
        • Emeritus and Retired Faculty
      • Staff (All)
      • Postdoctoral Researchers
      • Contact Us
    • About

      • Welcome from Chair
      • History
      • Strategic Plan
      • Biostatistics News
        • Student Stories
        • Faculty Stories
        • Alumni Stories
      • Contact Us
      • Open Positions
      • Harvey Co Chien Endowed Scholarship for Biostatistics
  • Environmental and Occupational Health
  • Epidemiology
  • Health Policy and Management
  • Human Genetics
  • Infectious Diseases and Microbiology
  • Academics
    • MS
    • MS - HDS concentration
    • MS - SCG concentration
    • PhD
    • Courses
      • Class Schedule
    • Student Handbooks & Forms
  • Research & Practice
    • Centers Overview
    • Faculty Research
    • Student Research
    • Consulting
    • Practice
  • Admissions & Aid
    • Admissions
    • Financial Aid
    • Frequently Asked Questions
    • Request Information
  • Seminars & Events
    • Biostatistics Events
    • Biostatistics Seminars
    • Biostatistics Dissertation Defenses
    • Conferences
      • ENAR Spring Meeting
      • JSM
  • Who's Who
    • Faculty (All)
      • Primary Faculty
      • Secondary Faculty
      • Adjunct Faculty
      • Emeritus and Retired Faculty
    • Staff (All)
    • Postdoctoral Researchers
    • Contact Us
  • About
    • Welcome from Chair
    • History
    • Strategic Plan
    • Biostatistics News
      • Student Stories
      • Faculty Stories
      • Alumni Stories
    • Contact Us
    • Open Positions
    • Harvey Co Chien Endowed Scholarship for Biostatistics
Apply Now

Biostatistics Dissertation Defenses

Instructions

Events that appear on this page are in the Sponsoring Department of "Biostatistics" with "Dissertation Defenses" as its Category.

To add or edit events on this page, go to http://publichealth.pitt.edu/submit

Upcoming Biostatistice Dissertation Defenses

No events to display
Event

Previous

Biostatistics Dissertation Defense

John Pleis - Mixtures of Discrete and Continuous Variables: Considerations for Dimension Reduction

  Email this event to me @

Wednesday 5/30 1:00PM - 3:00PM
7139 Public Health, Peterson Seminar Room

John Pleis of the Department of Biostatistics defends his dissertation on "Mixtures of Discrete and Continuous Variables: Considerations for Dimension Reduction".

Graduate faculty of the University and all other interested parties are invited to attend


ABSTRACT:

When working with high-dimensional data, we are often presented with statistical and computational challenges. Statistically, a major concern is model overfitting, and we wish to take advantage of an assumed low-dimensional data substructure to reduce the dimensionality of the problem. Because of the current scale of some of these types of problems, dimension reduction methods have been an active research area for several years and many different methods have been developed (e.g., regularization, matrix rank constraints, principal components).

Classical methods, such as principal components, have been utilized frequently but often rely on Gaussian distributional assumptions.  When this assumption is not tenable, such as when we have mixtures of different types of data, dimension reduction approaches are more limited; the body of existing research on the analysis of multiple types of data is not nearly as developed.  As a result, this dissertation we will focus on certain aspects of dimension reduction for mixtures of continuous and discrete data.  Of particular interest is Mardia’s hypothesis that the first m eigenvalues contain all the important variation for the multivariate data.  However, this hypothesis test is based on the assumption of multivariate normality.  At this point, we are not aware of methodological developments addressing such aspects of principal components analysis for mixtures of continuous and discrete data.  Thus, we will first focus on GLOM-based (general location model) methodology attributed to Olkin and Tate for deriving the joint probability distribution of continuous and discrete random variables as the product of conditional and marginal probability distributions.  While existing hypothesis testing approaches generally assume multivariate normality, this dissertation will alter that paradigm slightly by focusing on the finite k-component mixture of multivariate normal distributions.  Because of this previous multivariate Gaussian assumption, the distribution of the sample covariance matrix has been assumed to be a central Wishart distribution. If the multivariate data is assumed to have been generated from a finite k-component mixture of multivariate normal distributions, an initial focus of this dissertation is to determine the distribution of the sample covariance matrix under this scenario.  

For this effort, we initially examine the distribution of the sample variance in a univariate setting by assuming the data was generated from a finite k-component mixture of Gaussian distributions.  By using concepts from the field of complex analysis via inversion of the Laplace transform, we will show that the distribution of the sample variance is a mixture of gamma distributions with different scale parameters.  Because such a distribution can be challenging to work with, we apply an approach similar to the Satterthwaite method to approximate the mixture distribution with a classical gamma distribution with one scale parameter via matching of the first two moments. 

In a similar manner, we examine the distribution of the sample covariance matrix in a multivariate setting by assuming the data was generated from a finite k-component mixture of multivariate normal distributions.  Also, we will derive marginal distributions when the joint distribution of the data is a finite k-component mixture of multivariate Gaussian distributions.  Through the use of the Laplace transform and Jacobians of a matrix argument, we will show that the distribution of the sample covariance is a mixture of Wishart distributions with differing scale parameters.  Similar to the univariate setting, such a distribution can be challenging to work with; a multivariate extension of the Satterthwaite method via moment matching may be of interest.  However, because the number of means, variances, and covariances is larger than the number of unknowns a different approach is demonstrated.  By using scalar representations of the sample covariance matrix (e.g., determinant, trace), we will illustrate approximating the mixture of Wishart distributions with a Wishart distribution having one scale parameter.  Because the approximating distribution may have fractional degrees of freedom, we examine some considerations for simulating data from a Wishart distribution with fractional degrees of freedom.  Finally, we evaluate the adequacy of the approximation method through the use of matrix norms (e.g., 1-norm, 2-norm, Frobenius).  Concluding remarks will summarize the impact of these developments on Mardia’s hypothesis test of proportionate eigenvalues.

KEYWORDS:  mixture distribution; eigenvalues; high-dimensional; Wishart; fractional degrees of freedom
    

Enrolled Users:

Last Updated On Tuesday, October 9, 2018 by Valenti, Renee Nerozzi
Created On Tuesday, April 24, 2018

  • Return
  • Export

Questions and Submissions

The schedule of dissertation defenses in the Department of Biostatistics is maintained by...

Renee Valenti
renee.valenti@pitt.edu
412-624-3023

Share the details for your defense to ensure it is displayed here as well as on the department calendar, Pitt Public Health's calendar, Weekly Update, and LCD screens.

Defenses by department

Follow the links below to view defenses in each department:

  • BCHS defenses
  • Biostatistics defenses
  • EOH defenses
  • Epidemiology defenses
  • HPM defenses
  • Human Genetics defenses
  • IDM defenses
  • Pitt Public Health defenses (all)

More event series

  • FULL SCHOOL CALENDAR
  • LCD screens (scrolling events)
  • The Week Ahead 
  • Grand Rounds
  • Student organization events
  • Food For Thought
  • One Book, One Community
  • Dissertation defenses
  • CPH continuing education
  • Academic calendar
  • Graduation deadlines
  • Professional Development
  • Career Services events and
    Pitt Bridges
  • Governance Meetings
  • Alumni Events
Back to Top

Map-Parran Hall, Pittsburgh, PA

Pitt Public Health
130 De Soto Street
Pittsburgh, PA 15261 USA

Contact Us
Prospective Students
What is Public Health?
Why Pitt Public Health?
Academics
Student Life
Admissions

Governance
Delta Omega
Resources
For Employers
Open Positions

Update Your Profile

Technology Support
Technology Policies

Communications Tools
and Samples
CEPH
© 2019 by University of Pittsburgh Graduate School of Public Health

Login  |  Sitemap