Exact tests of goodness of fit for log-linear models for rates
McDonald, John W., Smith, Peter W.F. and Forster, Jonathan J. (1999) Exact tests of goodness of fit for log-linear models for rates. Biometrics, 55, (2), 620-624. (doi:10.1111/j.0006-341X.1999.00620.x).
Download
Full text not available from this repository.
Description/Abstract
We propose Metropolis–Hastings sampling methods for estimating the exact conditional p-value for tests of goodness of fit of log-linear models for mortality rates and standardized mortality ratios. We focus on two-way tables, where the required conditional distribution is a multivariate noncentral hypergeometric distribution with known noncentrality parameter. Two examples are presented: a 2 × 3 table, where the exact results, obtained by enumeration, are available for comparison, and a 9 × 7 table, where Monte Carlo methods provide the only feasible approach for exact inference.
| Item Type: | Article |
|---|---|
| ISSNs: | 0006-341X (print) |
| Related URLs: | |
| Subjects: | Q Science > QA Mathematics |
| Divisions: | University Structure - Pre August 2011 > School of Mathematics > Statistics |
| Item ID: | 29958 |
| Date Deposited: | 19 Mar 2007 |
| Last Modified: | 01 Jun 2011 04:20 |
| Contributors: | McDonald, John W. (Author) Smith, Peter W.F. (Author) Forster, Jonathan J. (Author) |
| Date: | 1999 |
| Status: | Published |
| Contact Email Address: | jwm@alcd.soton.ac.uk |
| URI: | http://eprints.soton.ac.uk/id/eprint/29958 |
Actions (login required)
 |
View Item |
