STANDARD ERROR OF FORECAST IN MULTIPLE REGRESSION: PROOF OF A USEFUL RESULT,
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)
Document Details
- Document Type
- Technical Report
- Publication Date
- Apr 01, 1970
- Accession Number
- AD0705906
Entities
People
- Joseph S. Desalvo
Organizations
- RAND Corporation