Table 4Estimates of the Effect of Education Beyond High School on Log of Earningsa

RCALRMLRML2
Nonpenalized main effects (p = q1 = 19, q2 = 21)
θ 0 6.565 ± 0.2366.609 ± 0.274—
θ 1 6.402 ± 0.2976.307 ± 0.348—
LATE0.164 ± 0.3650.302 ± 0.439—
Linear spline with 3 knots (p = q1 = 114, q2 = 122)
θ 0 6.665 ± 0.3466.608 ± 0.3106.582 ± 0.364
θ 1 6.269 ± 0.3826.263 ± 0.3656.256 ± 0.406
LATE0.396 ± 0.5350.346 ± 0.4800.326 ± 0.538
Linear spline with 9 knots (p = q1 = 213, q2 = 233)
θ 0 6.675 ± 0.3186.627 ± 0.3156.586 ± 0.351
θ 1 6.197 ± 0.3596.234 ± 0.3586.196 ± 0.400
LATE0.478 ± 0.5050.393 ± 0.4790.390 ± 0.533
Linear spline with 15 knots (p = q1 = 312, q2 = 344)
θ 0 6.684 ± 0.3106.625 ± 0.3106.584 ± 0.341
θ 1 6.180 ± 0.3606.226 ± 0.3596.195 ± 0.394
LATE0.504 ± 0.4980.400 ± 0.4760.389 ± 0.522

Abbreviation: LATE, local average treatment effect.

aEstimate ± 1.96 × SE. The number of regressors is indicated by p, q1 or q2 in the instrument PS, treatment regression, or model. RCAL denotes θ^1,RCAL, RM L denotes θ^1,RCAL, and RML2 denotes the variant where the nuisance parameters are estimated by refitting models with only the variables selected from the corresponding Lasso estimation.

From: Developing and Testing New Methods for Estimating Treatment Effectiveness in Observational Studies Using High-Dimension Data

Cover of Developing and Testing New Methods for Estimating Treatment Effectiveness in Observational Studies Using High-Dimension Data
Developing and Testing New Methods for Estimating Treatment Effectiveness in Observational Studies Using High-Dimension Data [Internet].
Tan Z, Gerhard T, Sun B.
Copyright © 2022. Rutgers, The State University of New Jersey. All Rights Reserved.

This book is distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License which permits noncommercial use and distribution provided the original author(s) and source are credited. (See https://creativecommons.org/licenses/by-nc-nd/4.0/

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.