Header

UZH-Logo

Maintenance Infos

Testing-Based Forward Model Selection


Kozbur, Damian (2018). Testing-Based Forward Model Selection. Working paper series / Department of Economics 283, University of Zurich.

Abstract

This paper introduces and analyzes a procedure called Testing-Based Forward Model Selection (TBFMS) in linear regression problems. This procedure inductively selects covariates that add predictive power into a working statistical model before estimating a final regression. The criterion for deciding which covariate to include next and when to stop including covariates is derived from a profile of traditional statistical hypothesis tests. This paper proves probabilistic bounds for prediction error and the number of selected covariates, which depend on the quality of the tests. The bounds are then specialized to a case with heteroskedastic data with tests derived from Huber-Eicker-White standard errors. TBFMS performance is compared to Lasso and Post-Lasso in simulation studies. TBFMS is then analyzed as a component into larger post-model selection estimation problems for structural economic parameters. Finally, TBFMS is used to illustrate an empirical application to estimating determinants of economic growth.

Abstract

This paper introduces and analyzes a procedure called Testing-Based Forward Model Selection (TBFMS) in linear regression problems. This procedure inductively selects covariates that add predictive power into a working statistical model before estimating a final regression. The criterion for deciding which covariate to include next and when to stop including covariates is derived from a profile of traditional statistical hypothesis tests. This paper proves probabilistic bounds for prediction error and the number of selected covariates, which depend on the quality of the tests. The bounds are then specialized to a case with heteroskedastic data with tests derived from Huber-Eicker-White standard errors. TBFMS performance is compared to Lasso and Post-Lasso in simulation studies. TBFMS is then analyzed as a component into larger post-model selection estimation problems for structural economic parameters. Finally, TBFMS is used to illustrate an empirical application to estimating determinants of economic growth.

Statistics

Downloads

25 downloads since deposited on 23 Apr 2018
9 downloads since 12 months
Detailed statistics

Additional indexing

Item Type:Working Paper
Communities & Collections:03 Faculty of Economics > Department of Economics
Working Paper Series > Department of Economics
Dewey Decimal Classification:330 Economics
JEL Classification:C55
Uncontrolled Keywords:Model selection, forward regression, sparsity, hypothesis testing, Modellwahl, Lineare Regression, Wahrscheinlichkeitsverteilung, Statistischer Test
Language:English
Date:April 2018
Deposited On:23 Apr 2018 15:52
Last Modified:27 Nov 2020 07:29
Series Name:Working paper series / Department of Economics
Number of Pages:66
ISSN:1664-7041
Additional Information:Revised version Auch erschienen in: arXiv: 1512.02666v6
OA Status:Green
Official URL:http://www.econ.uzh.ch/static/wp/econwp283.pdf
Related URLs:http://www.econ.uzh.ch/static/workingpapers.php

Download

Green Open Access

Download PDF  'Testing-Based Forward Model Selection'.
Preview
Content: Published Version
Filetype: PDF
Size: 1MB