haku: @author Chatterjee, S. / yhteensä: 28
viite: 21 / 28
| Tekijä: | Teebagy, N. Chatterjee, S. |
| Otsikko: | Inference in a binary response model with application to data analysis. |
| Lehti: | Decision Sciences
1989 : SPRING, VOL. 20:2, p. 393-403 |
| Asiasana: | LINEAR MODELS STATISTICAL METHODS DECISION THEORY |
| Kieli: | eng |
| Tiivistelmä: | Standard errors of the coefficient of a logistic regression (a binary response model) based on the asymptotic formula are compared to those given by the bootstrap through Monte Carlo simulation. The former method underestimates the true value, while the latter one overestimates it. However, for small samples the underestimation is much larger than the overestimation. Two illustrative data set is used for the discussion. The first example deals with the issue of passing the ERA amendments as a function of women legislators and the percentage vote for Regan. The second example refers to a metallurgical problem. |
SCIMA