 (1)-1.png){kind=logo}
Course : Econometrics
Course code : OIK438
| First Name | Size | Date | ||
|---|---|---|---|---|
|
|
A quite improper introduction to optimisation based estimators and related testing procedures (v. 24/01/23-contains several corrections and an appendix with endnotes)
The notes are in a status of perpetual correction. Please report any typos to stelios@aueb.gr or the course's e-class.
|
73 MB | 1/24/23, 11:33 PM | |
|
|
Consistency (Theory and Examples)(v. 24/01/23-contains several corrections and an appendix with endnotes)
The notes are in a status of perpetual correction. Please report any typos to stelios@aueb.gr or the course's e-class.
|
67.49 MB | 1/24/23, 11:46 PM | |
|
|
36.36 KB | 9/5/23, 7:13 PM | ||
|
|
41.56 KB | 2/13/23, 11:15 PM | ||
|
|
The estimator is actually the (Conditional MLE) given that the elements of the z process are independently drawn from the standard Gaussian distribution. If Gaussianity does not hold, then the estimator is termed Gaussian Quasi MLE. You can obtain it in the experiment specified by the code by drawing independently the zs from some other distribution with mean zero and unit variance, e.g. a student's t with degrees of freedom greater than 2, properly standardized so that it has unit variance. Try it and examine numerically the issue of consistency.
|
917 B | 1/24/23, 11:55 PM |