| Full text | |
| Author(s): |
Patriota, Alexandre G.
Total Authors: 1
|
| Document type: | Journal article |
| Source: | SANKHYA-SERIES B-APPLIED AND INTERDISCIPLINARY STATISTICS; v. 74, n. 1, p. 22-pg., 2012-05-01. |
| Abstract | |
In this paper we introduce a q-Exponential regression model for fitting data with discrepant observations. Maximum likelihood estimators for the model parameters and the (observed and expected) Fisher information matrix are derived. Moreover, we also present sufficient conditions to have consistent and asymptotically normally distributed estimators. Simulation studies show that the proposed estimators present good behavior in the sense of decreasing bias, and symmetric distribution when the sample size increases. (AU) | |