What linear estimators miss

The effects of family income on child outcomes

Katrine V. Løken, Magne Mogstad, Matthew Wiswall

Research output: Contribution to journalArticle

42 Citations (Scopus)

Abstract

We assess the implications of nonlinearity for IV and FE estimation when the estimated model is inappropriately assumed to be linear. Our application is the causal link between family income and child outcomes. Our nonlinear IV and FE estimates show an increasing, concave relationship between family income and children's outcomes. We find that the linear estimators miss the significant effects of family income because they assign little weight to the large marginal effects in the lower part of the income distribution. We also show that the linear IV and FE estimates differ primarily because of different weighting of marginal effects.

Original languageEnglish (US)
Pages (from-to)1-35
Number of pages35
JournalAmerican Economic Journal: Applied Economics
Volume4
Issue number2
DOIs
StatePublished - Apr 2012
Externally publishedYes

Fingerprint

Estimator
Family income
Marginal effects
Weighting
Nonlinearity
Income distribution

ASJC Scopus subject areas

  • Economics, Econometrics and Finance(all)

Cite this

What linear estimators miss : The effects of family income on child outcomes. / Løken, Katrine V.; Mogstad, Magne; Wiswall, Matthew.

In: American Economic Journal: Applied Economics, Vol. 4, No. 2, 04.2012, p. 1-35.

Research output: Contribution to journalArticle

Løken, Katrine V. ; Mogstad, Magne ; Wiswall, Matthew. / What linear estimators miss : The effects of family income on child outcomes. In: American Economic Journal: Applied Economics. 2012 ; Vol. 4, No. 2. pp. 1-35.
@article{d2459fbb457d4a6db33518afb5cdde4f,
title = "What linear estimators miss: The effects of family income on child outcomes",
abstract = "We assess the implications of nonlinearity for IV and FE estimation when the estimated model is inappropriately assumed to be linear. Our application is the causal link between family income and child outcomes. Our nonlinear IV and FE estimates show an increasing, concave relationship between family income and children's outcomes. We find that the linear estimators miss the significant effects of family income because they assign little weight to the large marginal effects in the lower part of the income distribution. We also show that the linear IV and FE estimates differ primarily because of different weighting of marginal effects.",
author = "L{\o}ken, {Katrine V.} and Magne Mogstad and Matthew Wiswall",
year = "2012",
month = "4",
doi = "10.1257/app.4.2.1",
language = "English (US)",
volume = "4",
pages = "1--35",
journal = "American Economic Journal: Applied Economics",
issn = "1945-7782",
publisher = "American Economic Association",
number = "2",

}

TY - JOUR

T1 - What linear estimators miss

T2 - The effects of family income on child outcomes

AU - Løken, Katrine V.

AU - Mogstad, Magne

AU - Wiswall, Matthew

PY - 2012/4

Y1 - 2012/4

N2 - We assess the implications of nonlinearity for IV and FE estimation when the estimated model is inappropriately assumed to be linear. Our application is the causal link between family income and child outcomes. Our nonlinear IV and FE estimates show an increasing, concave relationship between family income and children's outcomes. We find that the linear estimators miss the significant effects of family income because they assign little weight to the large marginal effects in the lower part of the income distribution. We also show that the linear IV and FE estimates differ primarily because of different weighting of marginal effects.

AB - We assess the implications of nonlinearity for IV and FE estimation when the estimated model is inappropriately assumed to be linear. Our application is the causal link between family income and child outcomes. Our nonlinear IV and FE estimates show an increasing, concave relationship between family income and children's outcomes. We find that the linear estimators miss the significant effects of family income because they assign little weight to the large marginal effects in the lower part of the income distribution. We also show that the linear IV and FE estimates differ primarily because of different weighting of marginal effects.

UR - http://www.scopus.com/inward/record.url?scp=84866950980&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84866950980&partnerID=8YFLogxK

U2 - 10.1257/app.4.2.1

DO - 10.1257/app.4.2.1

M3 - Article

VL - 4

SP - 1

EP - 35

JO - American Economic Journal: Applied Economics

JF - American Economic Journal: Applied Economics

SN - 1945-7782

IS - 2

ER -