Testing the Linear Regression Model Using Burr Critical Value and p-Value Approximations

The accuracy of Burr approximations of critical values and p-values is evaluated for the usual tests of the general linear regression model. These include tests for coefficients and autocorrelation and heteroscedasticity, based both on standard distributions and those for which the distribution is unknown. The results suggest that the Burr approximations are reasonably accurate and should prove useful both in applied research and in teaching.


Issue Date:
Dec 01 1990
Publication Type:
Working or Discussion Paper
Record Identifier:
http://ageconsearch.umn.edu/record/267145
Language:
English
Total Pages:
27
Series Statement:
Working Paper No. 20/90




 Record created 2018-01-24, last modified 2018-01-25

Fulltext:
Download fulltext
PDF

Rate this document:

Rate this document:
1
2
3
 
(Not yet reviewed)