The major challenges on the statistical analysis of microarray data are the limited availability of samples, large number of measured variables and the complexity of distribution of the data obtained (e.g., multimodal). These phenomena could be considered in Bayesian method, used Bayesian Mixture Model (BMM) methods and Bayesian Model Averaging (BMA) methods. Modeling of Bayesian Mixture Model Averaging (BMMA) for microarray data was developed based on these two studies. One of the most important stages in BMMA is determination of the number of mixture components in the data setting as the most appropriate BMMA models. This paper proposes an algorithm for determining the number of mixture components in BMMA for microarray data. The algorithm is developed based on the simulation data generated from a case study of Indonesian and it has been implemented on the outside Indonesian microarray data. The results have succed to demonstrate two step algorithms, called Preliminary Process and Smoothing Process Algorithms, to the Indonesian case microarray data with the accuracy rate of 99.3690% and 99.9094% for the outside Indonesian microarray data.

This study discusses the development of Zero Inflated Generalized Poisson Regression (ZIGPR) with two response variables, that is Bivariate ZIGPR (BZIGPR). The extension of the ZIGPR model by considering spatial factor called Geographically Weighted Zero Inflated Generalized Poisson Regression (GWBZIGPR). The GWBZIGPR produces a local parameter estimator for each location of observation. The parameter estimation using the Maximum Likelihood Estimation (MLE) method obtained an equation that did not closed-form so that the numerical iteration of Berndt Hall Hall Hausman (BHHH) is used. The data used in this study are the number of pregnant maternal mortality and postpartum maternal mortality data in 91 sub-districts in Pekalongan Residency, Central Java Province. The results showed that the Akaike Information Criterion Corrected (AICc) value in the GWBZIGPR model is smaller than BZIGPR, so it means that the GWBZIGPR is better than the BZIGPR for modeling the number of pregnant maternal mortality and postpartum maternal mortality in Pekalongan Residency. The results of this study will assist local governments in anticipating the causes of maternal mortality.

The Bayesian Model Averaging (BMA) required the validation step to determine the accuracy of BMA model. Kolmogorov-Smirnov (KS) and Continuous Ranked Probability Score (CRPS) are used to validate the BMA model. The absolute difference between the empirical cumulative distribution and the hypothesis cumulative distribution were the basic idea of these methods. The KS method uses the distance concept and CRPS method uses the area concept. The validation of BMA model on microarray data by KS and CRPS methods would be identified in this paper. The results have succeed to indentify the performance of KS and CRPS in the validation to BMA model on microarray data with an average value of KS=0.469 and CRPS=0.211 for n=10 and then the value of KS=0.403 and CRPS=0.11 for n=12.

No abstract

No abstract