Asymptotically Unbiased Inference for a Dynamic Panel Model with Fixed Effects when Both N and T are Large

By Jinyong Hahn, G. Kuersteiner. 2002

491 citations recorded by the source corpus.

We consider a dynamic panel AR(1) model with fixed effects when both "n" and "T" are large. Under the "T fixed n large" asymptotic approximation, the maximum likelihood estimator is known to be inconsistent due to the well-known incidental parameter problem. We consider an alternative asymptotic approximation where "n" and "T" grow at the same rate. It is shown that, although the MLE is asymptotically biased, a relatively simple fix to the MLE results in an asymptotically unbiased estimator. The bias corrected MLE is shown to be asymptotically efficient by a Hajek type convolution theorem.

Subfields: Economics and Econometrics

View Paper: https://www.semanticscholar.org/paper/8c0aea10a14d2175192e2cb31fbc14a8d4052d6c

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Economics and Econometrics

Citations in the corpus

Common Correlated Effects Estimation of Heterogeneous Dynamic Panel Data Models with Weakly Exogenous Regressors (2015, unknown): “…(i) analytical corrections based on an asymptotic bias formula (Bruno, 2005, Bun, 2003, Bun and Carree, 2005 and 2006, Bun and Kiviet, 2003, Hahn and Kuersteiner, 2002 and 2011, Hahn and Moon, 2006, Hahn and Newey, 2004, Kiviet, 1995 and 1999, and Newey and Smith, 2004); (ii) bootstrap and…”
Spatial Vector Autoregressions (2007, background): “‘For example, an intercountry comparison may well include the full set of countries for which it is reasonable to assume that the model is constant’ (Greene, 2003, p. 293). Matters would be different if the spatial units in the data set were a random sample of the spatial units in the population, su”
The Incidental Parameters Problem in Testing for Remaining Cross-Section Correlation (2018, methodology): “These circumstances are reminiscent of the panel data setup considered in, for example, Phillips and Moon (1999) or Hahn and Kuersteiner (2002), where estimation of many (incidental) parameters in linear regression models turns out to have distributional effects on the asymptotic properties of…”
Penalized Quantile Regression for Dynamic Panel Data (2010, background): “See, for example, Arellano and Hahn (2007) for a survey, and Hahn and Kuersteiner (2002, 2003), Alvarez and Arellano (2003), Hahn and Newey (2004), Arellano and Hahn (2006), Bester and Hansen (2009), for specific approaches.”
Bias-Corrected Estimation in Dynamic Panel Data Models (2005, methodology): “Finally, Hahn and Kuersteiner (2002) recently introduced a bias-corrected estimator related to that developed by Kiviet (1995); however, their estimator is not designed for samples with small T .”
Bias Corrected Instrumental Variables Estimation for Dynamic Panel Models with Fixed Effects (2001, unknown): “Using these arguments and Lemmas 12, 13, and 14 below, we may rewrite the first order condition (22) as”
Grouped effects estimators in fixed effects models (2016, methodology): “Consider the case ρ = 1, where θ̂FE is asymptotically biased, but the bias correction proposed by Hahn and Kuersteiner (2004) results in an asymptotically unbiased estimator.”
Time-Varying Risk Premium in Large Cross-Sectional Equity Data Sets (2016, unknown): “In the likelihood setting of Hahn and Newey (2004) (see also Hahn and Kuersteiner (2002)), the bis correspond to the individual fixed effects and ν to the common parameter of interest.”
Asymptotic local power of pooled t‐ratio tests for unit roots in panels with fixed effects (2008, unknown): “(2002) (LLC thereafter) who propose t-ratio type tests, Im et al. (2003) who propose average type tests, and Maddala and Wu (1999) and Choi (2001) who propose Fisher-type statistics.”
Bias in Dynamic Panel Estimation with Fixed Effects, Incidental Trends and Cross Section Dependence (2003, methodology): “, limρ→1G (ρ, T ) = − 3 T+1 , and the inconsistency of the pooled least squares estimate for ρ = 1 follows plimN→∞(ρ̂− 1) = − 3 T + 1 , (8) a result that can be confirmed by more tedious direct calculation for the case ρ = 1.”
THE MOVING BLOCKS BOOTSTRAP FOR PANEL LINEAR REGRESSION MODELS WITH INDIVIDUAL FIXED EFFECTS (2011, background): “If n and T are of comparable size and the regressors contain lagged dependent variables (as in Hahn and Kuersteiner, 2002; Alvarez and Arellano, 2003; for more recent papers that discuss bias correction in the context of panel models estimated with interactive fixed effects, see also Bai, 2009;…”
THE ASYMPTOTIC PROPERTIES OF THE SYSTEM GMM ESTIMATOR IN DYNAMIC PANEL DATA MODELS WHEN BOTH N AND T ARE LARGE (2015, background): “As shown in Hahn and Kuersteiner (2002), if we assume that vi t is normally distributed, N (0,1 −α2) is the minimal asymptotic distribution.”
Accuracy and Efficiency of Various GMM Inference Techniques in Dynamic Micro Panel Data Models (2017, result): “For example: the simulation results in Arellano and Bover [3], Hahn and Kuersteiner [4], Alvarez and Arellano [5], Hahn et al.”
DYNAMIC LINEAR PANEL REGRESSION MODELS WITH INTERACTIVE FIXED EFFECTS (2017, background): “Hahn & Kuersteiner (2002) introduced the alternative asymptotics to characterize the asymptotic bias due to incidental parameter problems in fixed effect dynamic panel data models.”
A SIMPLE EFFICIENT INSTRUMENTAL VARIABLE ESTIMATOR FOR PANEL AR(p) MODELS WHEN BOTH N AND T ARE LARGE (2009, unknown): “To prove (14), we have to show that g̃it is Op(1/ √ t).”
Generalized least squares inference in panel and multilevel models with serial correlation and fixed effects (2007, methodology): “Conclusions (i) and (ii) of Proposition 2 mirror similar results from Hahn and Kuersteiner (2002) and Hahn and Newey (2004), demonstrating that bias remains in the limiting distribution of α̂ even when T grows as fast as S, but that this bias is removed if the one-step bias correction is used as…”
Bias Corrections for Two-Step Fixed Effects Panel Data Estimators (2007, background): “¶ : ( 9 ) If T grows at the same rate as n the FE estimator, while consistent, has a limiting”