Comet Calendar

Statistics Seminar by Ashok Chaurasia (Waterloo)

Friday, December 9, 2022 11am to 12pm

Virtual Event

Model-modified BIC as a competitor of BIC variants for model selection in regression and order selection in time series

Model Selection is a ubiquitous task in the model building process that seeks a parsimonious model to explain patterns in data. Bayesian Information Criteria (BIC) is one such tool that is used in applications for its property of consistency, where this term is traditionally synonymous with asymptotic consistency. This article introduces the concept of exact consistency in model selection (which places an upper bound on the probability of selecting under-fitted models) to elucidate finite-sample properties of asymptotically consistent criteria. A new class of criteria gBIC is proposed with conditions for candidate model priors that guarantee exact and asymptotic consistency. gBIC provides a theoretical means to assess the consistency of existing BIC variants by simply checking functional properties of their respective penalty components. It is shown that gBIC includes sample size adjusted BIC (SABIC) as a special case, and improves on BIC with the proposed model-modified BIC (mBIC). In comparison to BIC, mBIC guarantees higher success rates for selecting models that match or contain the data generating model at any sample size for linear regression and time series. Lastly, mBIC can easily be adopted in linear regression and time series via simple algebraic adjustments to BIC/SABIC values.

Virtual Event

Natural Sciences & Mathematics
Qiwei Li
Email

UTD strives to create inclusive and accessible events in accordance with the Americans with Disabilities Act (ADA). If you require an accommodation to fully participate in this event, please contact the event coordinator (listed above) at least 10 business days prior to the event. If you have any additional questions, please email ADACoordinator@utdallas.edu and the AccessAbility Resource Center at accessability@utdallas.edu.

Event Publishing

Add an Event 

Submit your own event using our simple event submission form. It only takes a minute!

Event Publisher Training 

Learn best practices to maximize the calendar’s latest features.

Make a Calendar Feed 

Embed events anywhere on the web with our widget builder.

Explore Comet Calendar

Events by Email

Get a personalized list of events in your inbox with our digest emailer.

30-second Survey

Share your feedback and suggestions on how we can improve the Comet Calendar.

The University of Texas at Dallas