我們提出了一个混合迴歸模型下的貝氏變數選取法，此法可以自動適應模型的不確定性、母體的異質性以及離群值效應。變數選取借鑑資料增強的想法和特殊的先驗分配，且模型的推論主要通過馬爾科夫鏈蒙地卡羅演算法之結果實現。此模型被運用在分析全球金融危機資料上：在兩個次母體存在的前提下，我們找到各個次母體中的重要變數，以及一些可能的離群值。

A Bayesian variable selection approach for finite mixture regression model is proposed,which is able to simultaneously accommodate model uncertainty, population heterogeneity
and outlier effect. Variable selection is mainly accomplished through the idea of data augmentation and special spike and slab prior specification, and model inference is based on MCMC output. The proposed method is further applied to analyze the global financial crises data. Under two-subpopulation setting, some important covariates for each group are found, as well as several countries that are possible outliers.

Diebolt, J. and Robert, C. P. (1994). Estimation of finite mixture distributions through Bayesian sampling. Journal of the Royal Statistical Society. Series B (Methodological),
56(2):363–375.
Feldkircher, M., Horvath, R., and Rusnak, M. (2014). Exchange market pressures during the financial crisis: A Bayesian model averaging evidence. Journal of International
Money and Finance, 40(0):21 – 41.
George, E. I. and McCulloch, R. E. (1993). Variable selection via Gibbs sampling. Journal of the American Statistical Association, 88:881–889.
Geweke, J. (1993). Bayesian treatment of the independent Student-t linear model. Journal of Applied Econometrics, 8:S19–S40.
Green, P. J. (1995). Reversible jump Markov chain Monte Carlo computation and Bayesian model determination. Biometrika, 82:711–732.
Jasra, A., Holmes, C. C., and Stephens, D. A. (2005). Markov chain Monte Carlo methods and the label switching problem in Bayesian mixture modeling. Statistical Science,
20:50–67.
McLachlan, G. J. and Peel, D. (2000). Finite Mixture Models. Wiely, New York. Mitchell, T. J. and Beauchamp, J. J. (1988). Bayesian variable selection in linear regres-
sion. Journal of the American Statistical Association, 83(404):1023–1032.
O’Hara, R. B. and Sillanpää, M. J. (2009). A review of Bayesian variable selection methods: What, how and which. Bayesian Analysis, 4:85–118.
Redner, R. A. and Walker, H. F. (1984). Mixture densities, maximum likelihood and the em algorithm. SIAM Review, 26(2):195–239.
Tanner, M. A. and Wong, W. H. (1987). The calculation of posterior distributions by data augmentation. Journal of the American Statistical Association, 82(398):528–540.
Zellner, A. (1986). On assessing prior distributions and Bayesian regression analysis with g-prior distributions. In Goel, P. and Zellner, A., editors, Bayesian Inference and Decision Techniques: Essays in Honor of Bruno de Finetti, pages 233–243. North-Holland, Amsterdam.