Solving the omitted variables problem of regression analysis using the relative vertical position of observations

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

The omitted variables problem is one of regression analysis' most serious problems. The standard approach to the omitted variables problem is to find instruments, or proxies, for the omitted variables, but this approach makes strong assumptions that are rarely met in practice. This paper introduces best projection reiterative truncated projected least squares (BP-RTPLS), the third generation of a technique that solves the omitted variables problem without using proxies or instruments. This paper presents a theoretical argument that BP-RTPLS produces unbiased reduced form estimates when there are omitted variables. This paper also provides simulation evidence that shows OLS produces between 250% and 2450% more errors than BP-RTPLS when there are omitted variables and when measurement and round-off error is 1 percent or less. In an example, the government spending multiplier, ∂ GDP / ∂ G, is estimated using annual data for the USA between 1929 and 2010.

Original languageEnglish (US)
Article number728980
JournalAdvances in Decision Sciences
Volume2012
DOIs
StatePublished - Dec 1 2012

Fingerprint

Regression Analysis
Regression analysis
Vertical
Least Squares
Projection
Rounding error
Percent
Annual
Multiplier
Observation
Omitted variables
Estimate
Least squares
Simulation

ASJC Scopus subject areas

  • Decision Sciences(all)
  • Statistics and Probability
  • Computational Mathematics
  • Applied Mathematics

Cite this

@article{8b59502969a741c2b30e0a210b7101b7,
title = "Solving the omitted variables problem of regression analysis using the relative vertical position of observations",
abstract = "The omitted variables problem is one of regression analysis' most serious problems. The standard approach to the omitted variables problem is to find instruments, or proxies, for the omitted variables, but this approach makes strong assumptions that are rarely met in practice. This paper introduces best projection reiterative truncated projected least squares (BP-RTPLS), the third generation of a technique that solves the omitted variables problem without using proxies or instruments. This paper presents a theoretical argument that BP-RTPLS produces unbiased reduced form estimates when there are omitted variables. This paper also provides simulation evidence that shows OLS produces between 250{\%} and 2450{\%} more errors than BP-RTPLS when there are omitted variables and when measurement and round-off error is 1 percent or less. In an example, the government spending multiplier, ∂ GDP / ∂ G, is estimated using annual data for the USA between 1929 and 2010.",
author = "Leightner, {Jonathan E} and Tomoo Inoue",
year = "2012",
month = "12",
day = "1",
doi = "10.1155/2012/728980",
language = "English (US)",
volume = "2012",
journal = "Advances in Decision Sciences",
issn = "2090-3359",
publisher = "Hindawi Publishing Corporation",

}

TY - JOUR

T1 - Solving the omitted variables problem of regression analysis using the relative vertical position of observations

AU - Leightner, Jonathan E

AU - Inoue, Tomoo

PY - 2012/12/1

Y1 - 2012/12/1

N2 - The omitted variables problem is one of regression analysis' most serious problems. The standard approach to the omitted variables problem is to find instruments, or proxies, for the omitted variables, but this approach makes strong assumptions that are rarely met in practice. This paper introduces best projection reiterative truncated projected least squares (BP-RTPLS), the third generation of a technique that solves the omitted variables problem without using proxies or instruments. This paper presents a theoretical argument that BP-RTPLS produces unbiased reduced form estimates when there are omitted variables. This paper also provides simulation evidence that shows OLS produces between 250% and 2450% more errors than BP-RTPLS when there are omitted variables and when measurement and round-off error is 1 percent or less. In an example, the government spending multiplier, ∂ GDP / ∂ G, is estimated using annual data for the USA between 1929 and 2010.

AB - The omitted variables problem is one of regression analysis' most serious problems. The standard approach to the omitted variables problem is to find instruments, or proxies, for the omitted variables, but this approach makes strong assumptions that are rarely met in practice. This paper introduces best projection reiterative truncated projected least squares (BP-RTPLS), the third generation of a technique that solves the omitted variables problem without using proxies or instruments. This paper presents a theoretical argument that BP-RTPLS produces unbiased reduced form estimates when there are omitted variables. This paper also provides simulation evidence that shows OLS produces between 250% and 2450% more errors than BP-RTPLS when there are omitted variables and when measurement and round-off error is 1 percent or less. In an example, the government spending multiplier, ∂ GDP / ∂ G, is estimated using annual data for the USA between 1929 and 2010.

UR - http://www.scopus.com/inward/record.url?scp=84872816748&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84872816748&partnerID=8YFLogxK

U2 - 10.1155/2012/728980

DO - 10.1155/2012/728980

M3 - Article

VL - 2012

JO - Advances in Decision Sciences

JF - Advances in Decision Sciences

SN - 2090-3359

M1 - 728980

ER -