Estimation of dynamic econometric models with errors in variables (Q1188682): Difference between revisions
From MaRDI portal
Created a new Item |
Set profile property. |
||
(One intermediate revision by one other user not shown) | |||
Property / MaRDI profile type | |||
Property / MaRDI profile type: MaRDI publication profile / rank | |||
Normal rank | |||
links / mardi / name | links / mardi / name | ||
Latest revision as of 03:27, 5 March 2024
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Estimation of dynamic econometric models with errors in variables |
scientific article |
Statements
Estimation of dynamic econometric models with errors in variables (English)
0 references
17 September 1992
0 references
Econometricians are quite often confronted with errors in variables; be it that the variable of primary interest is not directly observable or that it is observed with measurement errors. Furthermore, it is known that regression models with measurement errors in the independent variables are not identified. To overcome these difficulties, the author presents a general procedure for the maximum likelihood estimation of dynamic econometric models with errors in both endogenous and exogenous variables. For this case the author considers the endogenous and exogenous variables of the original specification as observable variables in a new state-space parameterization. The first two chapters of the book (ignoring the introduction) deal with the formulation of dynamic econometric models in state-space with the extension of possible measurement errors in both types of variables. The center-piece of the book is chapter four, in which the author is concerned with the estimation problem. He constructs the likelihood function, analytical expressions of its gradient and Hessian. More technical aspects are given in seven appendices. In addition, the author compares possible advantages of the Chandrasekhar equations instead of the Riccati equations of the Kalman filter. Chapter five extends the previous analysis to situations with missing observations, to cases where only temporal aggregates of the variables are observed, and to the case of correlated measurement errors. A small simulation study informs the reader of the quality of the proposed estimation method. Chapter six contains the results. Chapter seven summarizes the main results of the book.
0 references
aggregation
0 references
numerical optimization
0 references
errors in variables
0 references
measurement errors
0 references
maximum likelihood estimation of dynamic econometric models
0 references
state-space parameterization
0 references
Chandrasekhar equations
0 references
Kalman filter
0 references
missing observations
0 references
simulation study
0 references