Abstract
Bayesian analysis is given of a random effects binary probit model that allows for heteroscedasticity. Real and simulated examples illustrate the approach and show that ignoring heteroscedasticity when it exists may lead to biased estimates and poor prediction. The computation is carried out by an efficient Markov chain Monte Carlo sampling scheme that generates the parameters in blocks. We use the Bayes factor, cross-validation of the predictive density, the deviance information criterion and Receiver Operating Characteristic (ROC) curves for model comparison.
| Language | English |
|---|---|
| Pages | 324-339 |
| Number of pages | 16 |
| Journal | Econometrics Journal |
| Volume | 12 |
| Issue number | 2 |
| DOIs | |
| Publication status | Published - 2009 |
| Externally published | Yes |
Fingerprint
Keywords
- Bayes factor
- Cross-validation
- Deviance information criterion
- Marginal effects
- Marginal likelihood
- Markov chain Monte Carlo
- ROC curve
Cite this
}
Bayesian estimation of a random effects heteroscedastic probit model. / Gu, Yuanyuan; Fiebig, Denzil G.; Cripps, Edward; Kohn, Robert.
In: Econometrics Journal, Vol. 12, No. 2, 2009, p. 324-339.Research output: Contribution to journal › Article › Research › peer-review
TY - JOUR
T1 - Bayesian estimation of a random effects heteroscedastic probit model
AU - Gu, Yuanyuan
AU - Fiebig, Denzil G.
AU - Cripps, Edward
AU - Kohn, Robert
PY - 2009
Y1 - 2009
N2 - Bayesian analysis is given of a random effects binary probit model that allows for heteroscedasticity. Real and simulated examples illustrate the approach and show that ignoring heteroscedasticity when it exists may lead to biased estimates and poor prediction. The computation is carried out by an efficient Markov chain Monte Carlo sampling scheme that generates the parameters in blocks. We use the Bayes factor, cross-validation of the predictive density, the deviance information criterion and Receiver Operating Characteristic (ROC) curves for model comparison.
AB - Bayesian analysis is given of a random effects binary probit model that allows for heteroscedasticity. Real and simulated examples illustrate the approach and show that ignoring heteroscedasticity when it exists may lead to biased estimates and poor prediction. The computation is carried out by an efficient Markov chain Monte Carlo sampling scheme that generates the parameters in blocks. We use the Bayes factor, cross-validation of the predictive density, the deviance information criterion and Receiver Operating Characteristic (ROC) curves for model comparison.
KW - Bayes factor
KW - Cross-validation
KW - Deviance information criterion
KW - Marginal effects
KW - Marginal likelihood
KW - Markov chain Monte Carlo
KW - ROC curve
UR - http://www.scopus.com/inward/record.url?scp=68149171594&partnerID=8YFLogxK
U2 - 10.1111/j.1368-423X.2009.00283.x
DO - 10.1111/j.1368-423X.2009.00283.x
M3 - Article
VL - 12
SP - 324
EP - 339
JO - Econometrics Journal
T2 - Econometrics Journal
JF - Econometrics Journal
SN - 1368-4221
IS - 2
ER -