We will compare several statistical package programs on aspects of exploratory data analysis(EDA). Specifically, PC versions of Minitab, NCSS, SAS, SPSS/PC+, Statgraphics, and SYSTAT are reviewed.

We consider the statistical control problem for the mixture model in which one can choose the values of independent variables that produce the values of the dependent variables as close to the target values as possible. The theory suggested for the problem is reviewed and an extended model with respect to the assumption of variance and the number of dependent variables is suggested. A Basyesian treatment is studied for the above problem with example as an illustration.

For the present study two sets of artificially unbalanced data of being $n_{ij}>0$ and were used. The Hypotheses that are commonly used in ANOVA were examined by computing the sums of squares associated with the hypotheses under various postulated models, using Searle's R()-notation.

In this paper we descuss some E-optimal block designs having unequal block sizes, and give a table of E-optimal designs with 2 different block sizes which can be constructed using the method described in Theorem 3. 2, Theorem 3. 4 and Theorem 3. 5 proved by Lee and Jacroux (1987). All of source designs used are Group Divisible designs which can be found in Clathworthy(1973) or Balanced Incomplete block designs in Raghavarar(1971).

For selecting a subset of k normal populations containing the one with the largest mean, a Bayes- selection procedure is considered when the common variance is unknown. Performance of the Bayes- selection procedure is compared with a well known classical procedure through a simulation study. Some frequentist's characteristics of Bayes- procedure are also studied.

Several practical prior distributions are derived using the maximum entropy principle. Also, an interactive method for estimating a prior distribution which uses the minimum cross-entropy principle is proposed when there are many prior informations. The consistency of the prior distributions obtained by the entropy principles is discussed.