Accession Number : ADA186033

Title :   Strategies of Data Analysis.

Descriptive Note : Technical rept.,

Corporate Author : PITTSBURGH UNIV PA CENTER FOR MULTIVARIATE ANALYSIS

Personal Author(s) : Rao, C R

PDF Url : ADA186033

Report Date : Jun 1987

Pagination or Media Count : 22

Abstract : The purpose of statistical analysis is 'to extract all the information from observed data'. The recorded data may have some detects such as recording errors and outliers and the first task of a statistician is to scrutinize or cross-examine the data for possible defects and understand its special features. The next step is the specification of a suitable stochastic model for the data using prior information and cross-validation techniques. On the basis of a chosen model, inferential analysis is made, which comprises of estimation of unknown parameters, tests of hypotheses, prediction of future observations and decision making. Examining data under different possible models is suggested as more informative than using robust procedures to safeguard against possible alternative models. Finally data analysis must also provide information for raising new questions and planning future investigations. Some aspects of data analysis as outlined above are illustrated through examples.

Descriptors :   *MULTIVARIATE ANALYSIS, DATA PROCESSING, DECISION MAKING, ERRORS, HYPOTHESES, MATHEMATICAL MODELS, STOCHASTIC PROCESSES, STRATEGY

Subject Categories : Statistics and Probability

Distribution Statement : APPROVED FOR PUBLIC RELEASE