1999 Fiscal Year Final Research Report Summary
Econometric method of economic data with measurement error and its application
| Project/Area Number |
10630019
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Economic statistics
|
|---|
| Research Institution | HOKKAIDO UNIVERSITY |
|---|
Principal Investigator |
HASEGAWA Hikaru Hokkaido Univ., Department of Economics, 経済学部, 助教授 (30189534)
|
|---|
| Project Period (FY) |
1998 – 1999
|
| Keywords | Measurement errors / Bayesian analysis / Engel curve / Markov chain Monte Carlo (MCMC) / HOGLEX demand system |
|---|
| Research Abstract |
In this research, we considered the Bayesian estimation of some economic models with the measurement errors on both the left and the right hand side. It is noteworthy that in the Bayesian approach no additional data are required in contrast with the non-Bayesian approach. In particular, we estimate the HOGLEX demand system and the Engel curves specified as Working-Leser form. We also derived the Bayesian estimation procedure in both models with and without an instrumental variable. Using household expenditure unit records from Thailand, Philippines and United Kingdom, this research applied the developed Bayesian methods to estimate the models. From the results of this research, the Bayesian methods are very suitable to the estimation of models with measurement errors and are applied to complex models, which are not estimated easily by the non-Bayesian methods.
|
