Estimating Distributed Lags in Short Panels with an Application to the Specification of Depreciation Patterns and Capital Stock Constructs (Q3339148)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 3872531
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | Estimating Distributed Lags in Short Panels with an Application to the Specification of Depreciation Patterns and Capital Stock Constructs |
scientific article; zbMATH DE number 3872531 |
Statements
Estimating Distributed Lags in Short Panels with an Application to the Specification of Depreciation Patterns and Capital Stock Constructs (English)
0 references
1984
0 references
constraints on the stochastic process
0 references
distributed lag model
0 references
time series
0 references
panel data
0 references
identification
0 references
capital stocks
0 references
rates of return
0 references
0.7867206931114197
0 references
0.7606697082519531
0 references
0.7606697082519531
0 references
0.7486766576766968
0 references
The simple distributed lag model for a single time series of observations is of the form \(y_ t=\alpha +\sum^{\infty}_{\tau =0}\omega_{\tau}x_{t-\tau}+\mu_ t,\) where conditional on \(\{x_ t\}\), the \(\mu_ t\) are independent draws from a common distribution function. With panel data for N individuals \((i=1,2,...,N)\) the model becomes \[ y_{i,t}=\alpha_ i+\sum^{\infty}_{\tau =0}\omega_{i,t-\tau}x_{i,t-\tau}+\mu_{it}. \] The assumption is made that differences in the lag coefficients among individuals are uncorrelated with their observed x sequences. This prior restriction allows one to combine all N time series into a model determined by a single sequence of lag coefficients: the mean of the lag distributions corresponding to different individuals in the population. For identification, two sorts of assumptions, e.g. constraints on the lag coefficients and on the stochastic processes generating the x sequences are investigated. Two examples which investigate how to empirically construct ''capital stocks'' for the analysis of rates of return are worked out.
0 references