Saved in:
Bibliographic Details
Main Authors: Farbmacher, Helmut, Groh, Rebecca, Mühlegger, Michael, Vollert, Gabriel
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2408.08580
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914000138141696
author Farbmacher, Helmut
Groh, Rebecca
Mühlegger, Michael
Vollert, Gabriel
author_facet Farbmacher, Helmut
Groh, Rebecca
Mühlegger, Michael
Vollert, Gabriel
contents Instrumental variables estimation with many instruments is biased. Traditional bias-adjustments are closely connected to the Silverstein equation. Based on the theory of random matrices, we show that Ridge estimation of the first-stage parameters reduces the implicit price of bias-adjustments. This leads to a trade-off, allowing for less costly estimation of the causal effect, which comes along with improved asymptotic properties. Our theoretical results nest existing ones on bias approximation and adjustment with ordinary least-squares in the first-stage regression and, moreover, generalize them to settings with more instruments than observations. Finally, we derive the optimal tuning parameter of Ridge regressions in simultaneous equations models, which comprises the well-known result for single equation models as a special case with uncorrelated error terms.
format Preprint
id arxiv_https___arxiv_org_abs_2408_08580
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Revisiting the Many Instruments Problem using Random Matrix Theory
Farbmacher, Helmut
Groh, Rebecca
Mühlegger, Michael
Vollert, Gabriel
Econometrics
Instrumental variables estimation with many instruments is biased. Traditional bias-adjustments are closely connected to the Silverstein equation. Based on the theory of random matrices, we show that Ridge estimation of the first-stage parameters reduces the implicit price of bias-adjustments. This leads to a trade-off, allowing for less costly estimation of the causal effect, which comes along with improved asymptotic properties. Our theoretical results nest existing ones on bias approximation and adjustment with ordinary least-squares in the first-stage regression and, moreover, generalize them to settings with more instruments than observations. Finally, we derive the optimal tuning parameter of Ridge regressions in simultaneous equations models, which comprises the well-known result for single equation models as a special case with uncorrelated error terms.
title Revisiting the Many Instruments Problem using Random Matrix Theory
topic Econometrics
url https://arxiv.org/abs/2408.08580