The bias of the RSR estimator and the accuracy of some alternatives

William N. Goetzmann, Liang Peng

Research output: Contribution to journalArticle

23 Citations (Scopus)

Abstract

This paper analyzes the implications of cross-sectional heteroskedasticity in the repeat sales regression (RSR). RSR estimators are essentially geometric averages of individual asset returns because of the logarithmic transformation of price relatives. We show that the cross-sectional variance of asset returns affects the magnitude of the bias in the average return estimate for each period, while reducing the bias for the surrounding periods. It is not easy to use an approximation method to correct the bias problem. We suggest an unbiased maximum likelihood alternative to the RSR that directly estimates index returns, which we term MLRSR. The unbiased MLRSR estimators are analogous to the RSR estimators but are arithmetic averages of individual asset returns. Simulations show that these estimators are robust to time-varying cross-sectional variance and that the MLRSR may be more accurate than RSR and some alternative methods.

Original languageEnglish (US)
Pages (from-to)13-39
Number of pages27
JournalReal Estate Economics
Volume30
Issue number1
DOIs
StatePublished - Jan 1 2002

Fingerprint

Estimator
Repeat sales
Asset returns
Relative prices
Approximation
Time-varying
Maximum likelihood
Simulation
Heteroskedasticity

All Science Journal Classification (ASJC) codes

  • Accounting
  • Finance
  • Economics and Econometrics

Cite this

@article{01cc0fb16dd84ac4b8cc79f198145d1f,
title = "The bias of the RSR estimator and the accuracy of some alternatives",
abstract = "This paper analyzes the implications of cross-sectional heteroskedasticity in the repeat sales regression (RSR). RSR estimators are essentially geometric averages of individual asset returns because of the logarithmic transformation of price relatives. We show that the cross-sectional variance of asset returns affects the magnitude of the bias in the average return estimate for each period, while reducing the bias for the surrounding periods. It is not easy to use an approximation method to correct the bias problem. We suggest an unbiased maximum likelihood alternative to the RSR that directly estimates index returns, which we term MLRSR. The unbiased MLRSR estimators are analogous to the RSR estimators but are arithmetic averages of individual asset returns. Simulations show that these estimators are robust to time-varying cross-sectional variance and that the MLRSR may be more accurate than RSR and some alternative methods.",
author = "Goetzmann, {William N.} and Liang Peng",
year = "2002",
month = "1",
day = "1",
doi = "10.1111/1540-6229.00028",
language = "English (US)",
volume = "30",
pages = "13--39",
journal = "Real Estate Economics",
issn = "1080-8620",
publisher = "Wiley-Blackwell",
number = "1",

}

The bias of the RSR estimator and the accuracy of some alternatives. / Goetzmann, William N.; Peng, Liang.

In: Real Estate Economics, Vol. 30, No. 1, 01.01.2002, p. 13-39.

Research output: Contribution to journalArticle

TY - JOUR

T1 - The bias of the RSR estimator and the accuracy of some alternatives

AU - Goetzmann, William N.

AU - Peng, Liang

PY - 2002/1/1

Y1 - 2002/1/1

N2 - This paper analyzes the implications of cross-sectional heteroskedasticity in the repeat sales regression (RSR). RSR estimators are essentially geometric averages of individual asset returns because of the logarithmic transformation of price relatives. We show that the cross-sectional variance of asset returns affects the magnitude of the bias in the average return estimate for each period, while reducing the bias for the surrounding periods. It is not easy to use an approximation method to correct the bias problem. We suggest an unbiased maximum likelihood alternative to the RSR that directly estimates index returns, which we term MLRSR. The unbiased MLRSR estimators are analogous to the RSR estimators but are arithmetic averages of individual asset returns. Simulations show that these estimators are robust to time-varying cross-sectional variance and that the MLRSR may be more accurate than RSR and some alternative methods.

AB - This paper analyzes the implications of cross-sectional heteroskedasticity in the repeat sales regression (RSR). RSR estimators are essentially geometric averages of individual asset returns because of the logarithmic transformation of price relatives. We show that the cross-sectional variance of asset returns affects the magnitude of the bias in the average return estimate for each period, while reducing the bias for the surrounding periods. It is not easy to use an approximation method to correct the bias problem. We suggest an unbiased maximum likelihood alternative to the RSR that directly estimates index returns, which we term MLRSR. The unbiased MLRSR estimators are analogous to the RSR estimators but are arithmetic averages of individual asset returns. Simulations show that these estimators are robust to time-varying cross-sectional variance and that the MLRSR may be more accurate than RSR and some alternative methods.

UR - http://www.scopus.com/inward/record.url?scp=0036522675&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036522675&partnerID=8YFLogxK

U2 - 10.1111/1540-6229.00028

DO - 10.1111/1540-6229.00028

M3 - Article

AN - SCOPUS:0036522675

VL - 30

SP - 13

EP - 39

JO - Real Estate Economics

JF - Real Estate Economics

SN - 1080-8620

IS - 1

ER -