Notes on bias in estimators for simultaneous equation models

By Jinyong Hahn, Jerry A. Hausman. 2002

135 citations recorded by the source corpus.

Subfields: Economics and Econometrics, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/3c5f37aa03a49b4e5631ebbd9526ce70346abebf

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

In Statistics and Probability

Citations in the corpus

Further results on projection-based inference in IV regressions with weak, collinear or missing instruments ∗ (2007, background): “For other results relevant to the instrument selection, the reader may consult Cragg and Donald (1993), Hall et al. (1996), Shea (1997), Chao and Swanson (2000), Donald and Newey (2001), Hall and Peixe (2003), Hahn and Hausman (2002a, 2002b) , and Stock and Yogo (2002).”
Estimation with Weak Instruments: Accuracy of Higher-order Bias and MSE Approximations (2004, unknown): “A partial list of such research activities is Nelson and Startz (1990), Maddala and Jeong (1992), Staiger and Stock (1997) and Hahn and Hausman (2002a). © Royal Economic Society 2004.”
Finite sample evidence of IV estimators under weak instruments (2007, background): “4 See also Hahn and Hausman (2002a) , who obtain a similar approximate finite sample bias expression for TSLS.”
The Issue of Endogeneity within Theory-Based, Quantitative Management Accounting Research (2007, background): “Although Hahn and Hausman (2002) indicate that the bias in the IV estimator is lower than the bias in the OLS estimator, the situation is more complicated when the instrument is not exogenous in the structural equation, that is, Cov(Z1, u) = 0.”
Estimation with Weak Instruments: Accuracy of Higher-Order Bias and MSE Approximations (2004, unknown): “2SLS has Þnite sample bias that depends on the number of instruments used K and inversely on the R2 of the Þrst stage regression, see e.g. Hahn and Hausman (2001).”
Selecting Instrumental Variables in a Data Rich Environment (2009, methodology): “Hahn and Hausman (2002) showed that the expected bias of 2SLS is linear in N , the number of instruments. The result arises because E(xPQε/T ) = E(uPQε/T ) = σuεN/T , where Q = (Q1, ..., QT )′ is T ×N . This motivates forming a set of L instruments with K2 ≤ L ≤ N by removing those elements of Qt th”
Instrumental Variables and GMM: Estimation and Testing (2003, background): “A concise online text that covers GMM is Hansen (2000).”
Improved JIVE Estimators for Overidentified Linear Models with and without Heteroskedasticity (2009, background): “Nagar’s estimator has recently been investigated by Hahn and Hausman (2002) and Donald and Newey (2001).”
Small sample bias properties of the system GMM estimator in dynamic panel data models (2007, background): “The second element of the bias results from using the first differencing and the level estimators jointly (the last row in eq(14)).”
The Weak Instrument Problem of the System GMM Estimator in Dynamic Panel Data Models (2010, unknown): “Hahn and Hausman (2002)) b which is equal to inconsistency of OLS.”
Enhanced routines for instrumental variables/GMM estimation and testing (2007, background): “The consequence of utilizing excluded instruments that are uncorrelated with the endogenous regressors is increased bias in the estimated IV coefficients (Hahn and Hausman (2002)) and worsening of the large-sample approximations to the finite-sample distributions.”
Estimation with Valid and Invalid Instruments (2003, unknown): “II Estimation with Valid Instruments From previous papers, e.g. Hahn and Hausman (2002a, 2002b) we know the bias and MSE of 2SLS up to second order.”
Are Exporters More Likely to Introduce Product Innovations (2012, unknown): “46, respectively (Stock and Yogo, 2005) 28We also estimated models using only one instrument at the time, since in the presence of weak instruments the size of the IVs bias is increasing in the number of instruments (Hahn and Hausman, 2002), but we obtained similar results without any gain in precis”
INSTRUMENTAL VARIABLE ESTIMATION IN A DATA RICH ENVIRONMENT (2010, methodology): “Hahn and Hausman (2002) showed that the expected bias of 2SLS is linear in r, the number of instruments. The result arises because E(xPF ε/T ) = E(uPF ε/T ) = σuεr/T . When F is not observed, we can still expect the bias of β̂FIV to increase with r because P̃ e F −PF = Op(min[ √ N, √ T ]−1) . This m”