Estimation for aggregate models: The aggregate Markov chain |
| |
Authors: | D. L. Mcleish |
| |
Affiliation: | Faculty of Mathematics Department of Statistics and Actuarial Science University of Waterloo Waterloo Ontario Canada N2L 3G1 |
| |
Abstract: | An argument in favour of projecting the score function for models involving incomplete data is presented. Projection is then applied to the aggregate Markov-chain model resulting in weighted least-squares estimators. The limit theory and efficiency of these estimators are studied using martingale limit theory. |
| |
Keywords: | Inference for stochastic processes Markov chains Estimation aggregate data least squares asymptotic efficiency Martingales estimating equations |
|