| |
|
Description: |
Likelihood-based inference for the Multi-state Latent Factor Intensity Model is hindered by the fact that exact closed-form expressions for the implied data density are not available. This is a common and well-known problem for most parameter driven dynamic Econometric models. The estimation of the MLFI model can be based on a combination of importance sampling and Kalman filtering techniques, as described in Durbin and Koopman (1997, 2000). Here, I review, adapt and compare two alternative solutions for solving the same problem. While the first method requires the use of Monte Carlo integration for evaluating the likelihood, the second method in contrast, is based on fully deterministic numerical procedures. A Monte Carlo study is conducted to illustrate the use of each method, and assess its corresponding finite sample performance.
|
|
Date: |
|
| Start Time: |
12:00 |
|
Speaker: |
André A. Monteiro (Tinbergen Institute, The Netherlands)
|
|
Institution: |
(Tinbergen Institute, The Netherlands)
|
| Research Groups: |
-Numerical Analysis and Optimization
|
|
See more:
|
<Main>
|
|
