Acknowledgements: I thank Harald Goldstein and Bernt P. Stigum for helpful
comments, which led to improvement in the organization of the paper. The usual
disclaimer applies.
Notes
1 Identification under non-normality of the true regressor is, however, possible by utilizing
moments of the distribution of the observable variables of order higher than the second [see
Reiers0l (1950)]. Even under non-identification, bounds on the parameters can be established
from the distribution of the observable variables [see Fuller (1987, p. 11)]. These bounds may
be wide or narrow, depending on the covariance structure of the variables; see, e.g., Klepper and
Leamer (1984) and Bekker et al. (1987).
2The last two assumptions are stronger than strictly needed; time invariance of E(αivit) and
E(αiuit) is sufficient. A modification to this effect will be of minor practical importance, however.
3 Premultiplication of (4) by dtθ is not the only way of eliminating αi .Any(1× T ) vector
ctθ such that ctθ eT = 0, for example the rows of the within individual transformation matrix
IT - eT eT0 /T , where IT is the T dimensional identity matrix, has this property.
4 Here and in the following plim always denotes probability limits when N goes to infinity and
T is finite.
5We report no standard error estimates in Table 24.1, since some of the methods are inconsis-
tent.
6 The OC’s involving y’s can be treated similarly. Essential and redundant moment conditions
in the context of AR models for panel data are discussed in, inter alia, Ahn and Schmidt (1995),
Arellano and Bover (1995), and Blundell and Bond (2000). This problem resembles, in some
respects, the problem for static measurement error models discussed here.
7 All numerical calculations are performed by means of procedures constructed by the author
in the GAUSS software code.
28
More intriguing information
1. AN ECONOMIC EVALUATION OF THE COLORADO RIVER BASIN SALINITY CONTROL PROGRAM2. Artificial neural networks as models of stimulus control*
3. Should Local Public Employment Services be Merged with the Local Social Benefit Administrations?
4. The name is absent
5. Word Sense Disambiguation by Web Mining for Word Co-occurrence Probabilities
6. Visual Perception of Humanoid Movement
7. School Effectiveness in Developing Countries - A Summary of the Research Evidence
8. Improving Business Cycle Forecasts’ Accuracy - What Can We Learn from Past Errors?
9. PROJECTED COSTS FOR SELECTED LOUISIANA VEGETABLE CROPS - 1997 SEASON
10. The name is absent