Multidimensional Parameter Heterogeneity in Panel Data Models

Timothy Neal

No 2016-15A, Discussion Papers from School of Economics, The University of New South Wales

Abstract: This article introduces a technique to estimate static or dynamic panel data models that feature two dimensions of heterogeneity in the slope and intercept parameters. It is able to consistently estimate the marginal effect for each individual observation as well as the average over a sample, and allows for correlation between the heterogeneity and the regressors. Models with two-dimensional fixed-effects in the slope parameters have long been considered interesting to economists yet intractable to estimate. Asymptotic theory establishes the consistency and asymptotic normality of the proposed estimator as N and T jointly go to infinity. Finally, Monte Carlo simulations demonstrate that the estimator performs well in environments where fixed effects and mean group estimators are inconsistent, inefficient, and severely biased.

Keywords: Estimation; Dynamic Modelling; Parameter Heterogeneity; Varying Coefficients (search for similar items in EconPapers)
JEL-codes: C13 C22 C23 C33 (search for similar items in EconPapers)
Pages: 32 pages
Date: 2018-02
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://research.economics.unsw.edu.au/RePEc/papers/2016-15A.pdf (application/pdf)

Related works:
Working Paper: Multidimensional Parameter Heterogeneity in Panel Data Models (2016)
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:swe:wpaper:2016-15a

Access Statistics for this paper

More papers in Discussion Papers from School of Economics, The University of New South Wales Contact information at EDIRC.
Bibliographic data for series maintained by Hongyi Li ().