Mixtures of g-priors for Bayesian Model Averaging with Economic Application
This paper examines the issue of variable selection in linear regression modeling, where there is a potentially large amount of possible covariates and economic theory offers insufficient guidance on how to select the appropriate subset. In this co...
| Main Authors: | , |
|---|---|
| Format: | Policy Research Working Paper |
| Language: | English |
| Published: |
2012
|
| Subjects: | |
| Online Access: | http://www-wds.worldbank.org/external/default/main?menuPK=64187510&pagePK=64193027&piPK=64187937&theSitePK=523679&menuPK=64187510&searchMenuPK=64187283&siteName=WDS&entityID=000158349_20110725090359 http://hdl.handle.net/10986/3496 |
| id |
okr-10986-3496 |
|---|---|
| recordtype |
oai_dc |
| spelling |
okr-10986-34962021-04-23T14:02:10Z Mixtures of g-priors for Bayesian Model Averaging with Economic Application Ley, Eduardo Steel, Mark F. J. ALGORITHMS ARCHIVE AREA ASPECT BAYES FACTOR BAYESIAN ANALYSIS BAYESIAN STATISTICS BAYESIAN THEORY BENCHMARK CALCULATION CLASSIFICATION COVARIANCE ECONOMETRICS ENTRIES ENUMERATION ESSAYS GAMMA DISTRIBUTION GENERALIZATION IDENTITY ILLUSTRATION INTEGRALS LINEAR MODELS LINEAR REGRESSION LITERATURE LITERATURES MATRIX MODELING MULTIPLE REGRESSION NESTED HYPOTHESES NORMAL DISTRIBUTIONS NOTATION POSTER PRECISION PREDICTION PREDICTIONS PROBABILITIES PROBABILITY RANDOM VARIABLES RANDOM WALK REASONING REGRESSION ANALYSIS SAMPLE SIZE SPIKE STANDARD DEVIATION STATISTICAL DECISION THEORY TERMINOLOGY UNION USER USERS VARIABILITY WEB This paper examines the issue of variable selection in linear regression modeling, where there is a potentially large amount of possible covariates and economic theory offers insufficient guidance on how to select the appropriate subset. In this context, Bayesian Model Averaging presents a formal Bayesian solution to dealing with model uncertainty. The main interest here is the effect of the prior on the results, such as posterior inclusion probabilities of regressors and predictive performance. The authors combine a Binomial-Beta prior on model size with a g-prior on the coefficients of each model. In addition, they assign a hyperprior to g, as the choice of g has been found to have a large impact on the results. For the prior on g, they examine the Zellner-Siow prior and a class of Beta shrinkage priors, which covers most choices in the recent literature. The authors propose a benchmark Beta prior, inspired by earlier findings with fixed g, and show it leads to consistent model selection. Inference is conducted through a Markov chain Monte Carlo sampler over model space and g. The authors examine the performance of the various priors in the context of simulated and real data. For the latter, they consider two important applications in economics, namely cross-country growth regression and returns to schooling. Recommendations for applied users are provided. 2012-03-19T18:03:29Z 2012-03-19T18:03:29Z 2011-07-01 http://www-wds.worldbank.org/external/default/main?menuPK=64187510&pagePK=64193027&piPK=64187937&theSitePK=523679&menuPK=64187510&searchMenuPK=64187283&siteName=WDS&entityID=000158349_20110725090359 http://hdl.handle.net/10986/3496 English Policy Research working paper ; no. WPS 5732 CC BY 3.0 IGO http://creativecommons.org/licenses/by/3.0/igo/ World Bank Publications & Research :: Policy Research Working Paper The World Region The World Region |
| repository_type |
Digital Repository |
| institution_category |
Foreign Institution |
| institution |
Digital Repositories |
| building |
World Bank Open Knowledge Repository |
| collection |
World Bank |
| language |
English |
| topic |
ALGORITHMS ARCHIVE AREA ASPECT BAYES FACTOR BAYESIAN ANALYSIS BAYESIAN STATISTICS BAYESIAN THEORY BENCHMARK CALCULATION CLASSIFICATION COVARIANCE ECONOMETRICS ENTRIES ENUMERATION ESSAYS GAMMA DISTRIBUTION GENERALIZATION IDENTITY ILLUSTRATION INTEGRALS LINEAR MODELS LINEAR REGRESSION LITERATURE LITERATURES MATRIX MODELING MULTIPLE REGRESSION NESTED HYPOTHESES NORMAL DISTRIBUTIONS NOTATION POSTER PRECISION PREDICTION PREDICTIONS PROBABILITIES PROBABILITY RANDOM VARIABLES RANDOM WALK REASONING REGRESSION ANALYSIS SAMPLE SIZE SPIKE STANDARD DEVIATION STATISTICAL DECISION THEORY TERMINOLOGY UNION USER USERS VARIABILITY WEB |
| spellingShingle |
ALGORITHMS ARCHIVE AREA ASPECT BAYES FACTOR BAYESIAN ANALYSIS BAYESIAN STATISTICS BAYESIAN THEORY BENCHMARK CALCULATION CLASSIFICATION COVARIANCE ECONOMETRICS ENTRIES ENUMERATION ESSAYS GAMMA DISTRIBUTION GENERALIZATION IDENTITY ILLUSTRATION INTEGRALS LINEAR MODELS LINEAR REGRESSION LITERATURE LITERATURES MATRIX MODELING MULTIPLE REGRESSION NESTED HYPOTHESES NORMAL DISTRIBUTIONS NOTATION POSTER PRECISION PREDICTION PREDICTIONS PROBABILITIES PROBABILITY RANDOM VARIABLES RANDOM WALK REASONING REGRESSION ANALYSIS SAMPLE SIZE SPIKE STANDARD DEVIATION STATISTICAL DECISION THEORY TERMINOLOGY UNION USER USERS VARIABILITY WEB Ley, Eduardo Steel, Mark F. J. Mixtures of g-priors for Bayesian Model Averaging with Economic Application |
| geographic_facet |
The World Region The World Region |
| relation |
Policy Research working paper ; no. WPS 5732 |
| description |
This paper examines the issue of
variable selection in linear regression modeling, where
there is a potentially large amount of possible covariates
and economic theory offers insufficient guidance on how to
select the appropriate subset. In this context, Bayesian
Model Averaging presents a formal Bayesian solution to
dealing with model uncertainty. The main interest here is
the effect of the prior on the results, such as posterior
inclusion probabilities of regressors and predictive
performance. The authors combine a Binomial-Beta prior on
model size with a g-prior on the coefficients of each model.
In addition, they assign a hyperprior to g, as the choice of
g has been found to have a large impact on the results. For
the prior on g, they examine the Zellner-Siow prior and a
class of Beta shrinkage priors, which covers most choices in
the recent literature. The authors propose a benchmark Beta
prior, inspired by earlier findings with fixed g, and show
it leads to consistent model selection. Inference is
conducted through a Markov chain Monte Carlo sampler over
model space and g. The authors examine the performance of
the various priors in the context of simulated and real
data. For the latter, they consider two important
applications in economics, namely cross-country growth
regression and returns to schooling. Recommendations for
applied users are provided. |
| format |
Publications & Research :: Policy Research Working Paper |
| author |
Ley, Eduardo Steel, Mark F. J. |
| author_facet |
Ley, Eduardo Steel, Mark F. J. |
| author_sort |
Ley, Eduardo |
| title |
Mixtures of g-priors for Bayesian Model Averaging with Economic Application |
| title_short |
Mixtures of g-priors for Bayesian Model Averaging with Economic Application |
| title_full |
Mixtures of g-priors for Bayesian Model Averaging with Economic Application |
| title_fullStr |
Mixtures of g-priors for Bayesian Model Averaging with Economic Application |
| title_full_unstemmed |
Mixtures of g-priors for Bayesian Model Averaging with Economic Application |
| title_sort |
mixtures of g-priors for bayesian model averaging with economic application |
| publishDate |
2012 |
| url |
http://www-wds.worldbank.org/external/default/main?menuPK=64187510&pagePK=64193027&piPK=64187937&theSitePK=523679&menuPK=64187510&searchMenuPK=64187283&siteName=WDS&entityID=000158349_20110725090359 http://hdl.handle.net/10986/3496 |
| _version_ |
1764387095108911104 |