
Accession Number : AD0705906
Title : STANDARD ERROR OF FORECAST IN MULTIPLE REGRESSION: PROOF OF A USEFUL RESULT,
Corporate Author : RAND CORP SANTA MONICA CALIF
Personal Author(s) : DeSalvo,Joseph S.
Report Date : APR 1970
Pagination or Media Count : 8
Abstract : Proof is given that the standard errors of forecasting the dependent variable or the expected value of the dependent variable in a multiple regression reduce to very simple formulas when evaluated at the sample means of the independent variables. These simple formulas involve only knowledge of sample size and the standard error of estimate, the latter of which is typically printed out in computer regression routines. By using these results, one avoids the necessity of calculating the more complicated general formulas for the standard errors in those cases for which evaluation at the mean will suffice. (Author)
Descriptors : (*REGRESSION ANALYSIS, MATHEMATICAL PREDICTION), MATHEMATICAL MODELS, MULTIPLE OPERATION, COMPUTER PROGRAMMING, SAMPLING, ERRORS
Subject Categories : Statistics and Probability
Distribution Statement : APPROVED FOR PUBLIC RELEASE