|
|||||
|
|
Hypothesis testing in regression models with AR(1) errors and a lagged dependent variable: Bootstrapping and artificial regression compared |
|
| Institution: | 1. Nova School of Business and Economics, Rua Holanda 1, Carcavelos 2775-405, Portugal;2. Federal Reserve Bank of New York, 33 Liberty Street, New York, NY 10045, USA;3. Department of Finance, Tilburg School of Economics and Management, LE Tilburg 5000, the Netherlands;4. Department of Finance, Rotterdam School of Management, Erasmus University, DR Rotterdam 3000, the Netherlands |
| Abstract: | Serially correlated errors in dynamic models render the standard conditional estimator of the covariance matrix inconsistent. A Monte Carlo experiment confirms that the downward bias in the conventional variance estimator also exists in small samples. The results favour a consistent estimator based on an artificial regression (suggested by Davidson and Mackinnon) over bootstrapping the distribution of parameter estimates. |
| Keywords: | |
| 本文献已被 ScienceDirect 等数据库收录！ | |