| RUS ENG |
| Полная версия |
|
|
|
| ВИДЕОТЕКА |
|
Workshop “Frontiers of High Dimensional Statistics, Optimization, and Econometrics”
|
|||
|
|
|||
|
[Linear hypothesis testing in the problem with weak instrumental variables] K. Andzhey |
|||
|
Аннотация: In this work a new method for testing linear hypothesis on the target parameter is developed. To compute critical values for a log-likelihood ratio statistics we use miltiplier bootstrap procedure. We justify this method and then use it for a regression model with instrumental varibles included. Numerical experiments show good power properties of proposed method under both weak and strong instruments identification. Язык доклада: английский |
|||