This paper proposes a test for the slope homogeneity in large dimensional panel data models with interactive fixed effects based on a measure of goodness-of-fit (R2). We first obtain, for each cross-sectional unit, the R2 from the time series regression of residuals on the constant and observable regressors and then construct the test statistic R2 as an equally weighted average of the cross-sectional R2's. R̄2 is close to 0 under the null hypothesis of homogenous slopes and deviates away from 0 otherwise. We show that after being appropriately centered and scaled, R2 is asymptotically normally distributed under the null and a sequence of Pitman local alternatives. To improve the finite sample performance of the test, we also propose a bootstrap procedure to obtain the bootstrap p-values and justify its validity. Monte Carlo simulations suggest that the test has correct size and satisfactory power, and is superior to a recent test proposed by Pesaran and Yamagata (2008) that neglects cross-sectional dependence in panel data models. We apply our tests to study the OECD economic growth model and the Fama French three factor model for asset returns.
cross-sectional dependence, goodness-of-fit, heterogeneity, interactive fixed effects, large panels, principal component analysis
MSc in Economics
Testing Heterogeneity in Panel Data Models with Interactive Fixed Effects. (2011). Dissertations and Theses Collection (Open Access).
Available at: https://ink.library.smu.edu.sg/etd_coll/76
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.