In this paper, we propose GroupAdaBoost which is a variant of AdaBoost for statistical pattern recognition. The objective of the proposed algorithm is to solve the “ p » n ”problem arisen in bioinformatics. In a microarray experiment, gene expressions are observed to extract any specific pattern of gene expressions related to a disease status. Typically, p is the number of investigated genes and n is number of individuals. The ordinary method for predicting the genetic causes of diseases is apt to over-learn from any particular training dataset because of the“ p » n ” problem. We observed that GroupAdaBoost gave a robust performance for cases of the excess number p of genes. In several real datasets which are publicly available from web-pages, we compared the analysis of results among the proposed method and others, and a small scale of simulation study to confirm the validity of the proposed method. Additionally the proposed method effectively worked for the identification of important genes.