This report introduces a new computer program, so-called , which has been developed at the Dalat Nuclear Research Institute, for data processing and interpretation of the experimental data sheets based on the multivariate data analysis techniques. In this preliminary version of the program, the dimensions of a given data set to be analyzed are up to 50 variables and thousands of observations. The main functions in this version are principal component analysis, cluster analysis, standardization and output data plot. | Communications in Physics, Vol. 27, No. 4 (2017), pp. 291-300 DOI: DEVELOPMENT OF A PC PROGRAM FOR MULTIVARIATE STATISTICAL ANALYSIS PHAM NGOC SON † , CAO DONG VU AND MAI QUYNH ANH Nuclear Research Institute, 1 Nguyen Tu Luc, Dalat, Vietnam † E-mail: Received 7 July 2017 Accepted for publication 11 September 2017 Published 18 November 2017 Abstract. This report introduces a new computer program, so-called , which has been developed at the Dalat Nuclear Research Institute, for data processing and interpretation of the experimental data sheets based on the multivariate data analysis techniques. In this preliminary version of the program, the dimensions of a given data set to be analyzed are up to 50 variables and thousands of observations. The main functions in this version are principal component analysis, cluster analysis, standardization and output data plot. In comparison with other well-known statistical analysis software programs, the same results are very well reproduced with . The format of the input data file was designed in a way convenient for the management of experimental survey data or preparation from other analysis procedures at the Institute. Keywords: multivariate data analysis, principal component analysis, cluster analysis. Classification numbers: , . I. INTRODUCTION In the modern trend of survey data, multi-dimension information is often observed in many cases of research and application. In these situations, the data reduction process or multivariate statistical analysis method should be performed in order to determine whether any distinct groups or distributions are present in the input data sheet that supports a meaningful interpretation in the research topic such as archeological, geological, environmental, etc. Based upon a data set of elemental concentrations, for a particular specimen, questions that we hope to answer are: (i) Where did the specimen come
