抄録
The clinical evaluation of medical treatments has been done on the basis of clinical trials with experimental designs or exploratory observations gathered. Generally spearking, such clinical evaluations are not so easy because of the complicated situations due to large variety and variation of observations, definition of the population and jointly the ethical point of view.The present paper presents some of the recent works developed by the author and his research colleagues.The first topic is related to the experimental trials, and is to propose a new the extended sequential selection plans based on generalized Play-the-Winner sampling. The plans proposed in a sence of the ethical situation are designed to give minimization of maximum expected loss during and after the medical trials for a finite populatino of patients. The most advantage of the procedure is to be much superior to the ordinal Vector-at-a-Time samling procedure on some major properties of selection plans, namelym, on probablilities of correct selection and wrong selection, expected number of subjects to select one and minimization of maximum expected loss mentioned above.The second is related to a new statistical program package, NISAN system, which is n__-ew i__-nteractive s__-tatistical <an>___-alysis program package constructed by an organization of Japanese statisticians, where author is representative. The package is widely available for both statistical situations, confirmatory analysis and exploratory analysis, and is planned to obtain statistical wisdom and to choose optimal process of statistical analysis for senior statisticians. In view of statistical methodology, the most emphasized functions are as follows; 1) data investigations, 2) graphic representations, 3) generalized methods of optimal scaling, 4) various methods of cluster analysis and multidimensional scaling, and 5) studies on successive processes of statistical inference.
