Split-Sample Instrumental Variables Estimates of the Return to Schooling

By Joshua D. Angrist, Alan B. Krueger. 1995

How researchers see this paper

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

Citations in the corpus

Double/Debiased Machine Learning for Treatment and Causal Parameters (2016, methodology): “al. (2012) who considered sample-splitting in a high-dimensional sparse optimal IV model to weaken the sparsity condition mentioned in the previous paragraph to s˝n. This work in turn was inspired by Angrist and Krueger (1995). We also build on Ayyagari (2010) and Robins et al. (2013), where ML meth”
Robust, accurate confidence intervals with a weak instrument: quarter of birth and education (2005, methodology): “Various improvements have been proposed, including the alternative asymptotics of Bekker (1994) and Staiger and Stock (1996), the jackknife methods of Angrist and Krueger (1995) and Angrist et al. (1999), the tests based Robust, Accurate Confidence Intervals 113”
Fertility and Female Labor Supply in Latin America: New Causal Evidence (2007, unknown): “…second panel in Table II we report the results obtained by split sample IV (SSIV) estimation, which are biased towards zero and not towards OLS (see Angrist and Krueger, 1995).8 The estimates obtained by SSIV are all significantly different from zero, and they are even higher (in absolute value)…”
Bootstrap Tests for Distributional Treatment Effects in Instrumental Variable Models (2002, methodology): “The data set was especially prepared for Angrist and Krueger (1995).”
Finite Sample Limited Information Inference Methods for Structural Equations and Models With Generated Regressors (2001, methodology): “A sample-split technique has also been suggested by Angrist and Krueger (1995) to build a new IV estimator, called split sample instrumental variables (SSIV) estimator.”
Double/debiased machine learning for treatment and structural parameters (2017, unknown): “This work in turn was inspired by Angrist and Krueger (1995). We also build on Ayyagari (2010)”
When to Control for Covariates? Panel Asymptotics for Estimates of Treatment Effects (2004, methodology): “The panel-asymptotic sequence is similar to sequences used by Bekker (1994), Bekker and van der Ploeg (1996), and Angrist and Krueger (1995) to analyze the finite-sample behavior of instrumental variables estimators.”
Sparse Models and Methods for Optimal Instruments With an Application to Eminent Domain (2012, methodology): “…that is related to Anderson and Rubin (1949) and Staiger and Stock (1997) but is better suited to cases with very many instruments; and we consider a split sample IV estimator in Section 5 which combines instrument selection via Lasso with the sample-splitting method of Angrist and Krueger (1995).”
Bias due to participant overlap in two‐sample Mendelian randomization (2016, background): “This approach is also known as split-sample 2SLS (Angrist & Krueger, 1995).”