Pivotal statistics for testing structural parameters in instrumental variables regression

By F. Kleibergen. 2002

509 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/b3e0bdc40d17063f9f6a908529ecb14942afc422

How researchers see this paper

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

In Statistics and Probability

Citations in the corpus

A MONTE CARLO COMPARISON OF VARIOUS ASYMPTOTIC APPROXIMATIONS TO THE DISTRIBUTION OF INSTRUMENTAL VARIABLES ESTIMATORS (2002, methodology): “We found that confidence intervals based on the Andersen and Rubin([12]) statistic and on the K statistic of Kleibergen([13]) have virtually exact Monte Carlo coverage probabilities: Their coverage probabilities were always within 1% point of the nominal coverage probability.”
Avoiding Invalid Instruments and Coping with Weak Instruments (2006, methodology): “To estimate this elasticity, economists typically specify a linear relationship between the rate of growth in consumption and the y Michael P. Murray is the Charles Franklin Phillips Professor of Economics, Bates College, Lewiston, Maine.”
Estimation With Many Instrumental Variables (2006, methodology): “We take as a standard of comparison our version of the Kleibergen (2002) confidence interval (denoted K in the graphs), which is robust to weak instruments, many instruments, and many weak instruments.”
Moving Average-Based Estimators of Integrated Variance (2008, background): “…ti ,m is the time of a transaction; while in the second (for which they reserve the term, tick time sampling)ti ,m is the time of a quote revision. persistent time series, see, e.g., Cavanagh et al. (1995), and the “weak instrument” asymptotics, see, e.g., Kleibergen (2002) and Moreira (2003).”
Robust Inference for Consumption‐Based Asset Pricing (2019, unknown): “In Kleibergen, Kong, and Zhan (2019), extensions of other weak identification-robust tests, like, the Lagrange multiplier and likelihood ratio tests (see, e.g., Kleibergen (2002, 2005) and Moreira (2003)) to this setting are discussed.”
Testing with many weak instruments (2007, methodology): “Such tests include the Anderson and Rubin (1949) (AR) test, the Lagrange multiplier (LM) test introduced in Kleibergen (2002) and Moreira (2001), and the conditional likelihood ratio (CLR) test introduced in Moreira (2003).”
On the asymptotic sizes of subset Anderson-Rubin and Lagrange multiplier tests in linear instrumental variables regression (2012, methodology): “Recently, other tests were introduced that improve on the power properties, in particular, the LM test, Kleibergen (2002) and the CLR test, Moreira (2003). The purpose of this section is to show that the subset version of the LM test, Kleibergen (2004), su¤ers from asymptotic size distortion.”
The Hausman test, and some alternatives, with heteroskedastic data (2012, background): “Inoue (2006) and Kleibergen (2002) present experiments which also indicate that the bootstrap can lead to size improvements for the irrelevant instruments case in the GMM context.”
Empirical Evidence on Inflation Expectations in the New Keynesian Phillips Curve (2014, background): “49Using weak identification robust stability tests, Kleibergen and Mavroeidis (2009) find some evidence of instability before 1984 when they use a shorter pre-1984 sample, but no evidence of instability using their full 1960-2008 sample, which is consistent with Magnusson and Mavroeidis (2012).”
GMM with many weak moment conditions (2005, methodology): “Further displayed is the p-value plot for Kleibergen’s (2005) K statistic.”
ASYMPTOTIC DISTRIBUTION OF JIVE IN A HETEROSKEDASTIC IV REGRESSION WITH MANY INSTRUMENTS (2011, methodology): “For rn = n, we have asymptotic theory as in Kunitomo (1980), Morimune (1984), and Bekker (1994), where the number of instruments Kn can grow as fast as the sample size.”
Frugal IV alternatives to identify the parameter for an endogenous regressor (2009, methodology): “Besides, several fully robust methods are developed to test hypothesis and construct confidence sets for fi\ that have approximately the correct size and coverage rates when instruments are weak (e.g., Anderson and Rubin, 1949; Kleibergen, 2002).”
Instrumental variable quantile regression: A robust inference approach (2008, methodology): “Quantile regression is an important tool for estimating conditional quantile models that has been used in many empirical studies and has been studied extensively in theoretical statistics; see Koenker and Bassett (1978), Koenker and Portnoy (1987), Portnoy (1991), Gutenbrunner and Jurečková (1992), ”
Unexplained factors and their effects on second pass R-squared’s (2015, unknown): “which converges to λF β PMιN βdβλF λF β MιN βλF , (21)”
The validity of instruments revisited (2012, background): “We also find that the Kleibergen (2002) test may not be amenable to the resampling technique that we use.”
Conditional Linear Combination Tests for Weakly Identified Models (2016, methodology): “For comparison, we consider Kleibergen (2005)’s quasi-CLR test, which takes r (D) = D′Σ−1 D D and can be viewed as a heteroskedasticity and autocorrelation-robust version of the CLR test, as well as the K and Anderson Rubin (S) statistics.”