Combining evidence from several samples for testing goodness-of-fit to a location-scale family. Technical report No. 15
Consider the problem of testing goodness-of-fit to a specified location-scale family when evidence is to be combined from several independent samples, from populations with possibly different location and scale parameters. The procedure studied is that of computing standardized residuals from each sample and then combining these into one set to be treated essentially as though they came from one sample. It is shown that the limiting distribution of any location-scale invariant goodness-of-fit statistic so applied is precisely the same as for the corresponding one-sample problem. 13 references.
- Research Organization:
- Stanford Univ., CA (USA). Dept. of Statistics
- Sponsoring Organization:
- USDOE
- OSTI ID:
- 6532602
- Report Number(s):
- COO-2874-35
- Country of Publication:
- United States
- Language:
- English
Similar Records
Testing goodness-of-fit for the distribution of errors in regression models. Technical report No. 16
A measure of the goodness of fit in unbinned likelihood fits
On Multivariate Goodness-of-Fit and Two-Sample Testing
Technical Report
·
Thu Jun 08 00:00:00 EDT 1978
·
OSTI ID:6532594
A measure of the goodness of fit in unbinned likelihood fits
Journal Article
·
Mon Jul 01 00:00:00 EDT 2002
· Phys.Rev.
·
OSTI ID:1155672
On Multivariate Goodness-of-Fit and Two-Sample Testing
Technical Report
·
Sun Jan 25 19:00:00 EST 2004
·
OSTI ID:826696