Forecast Accuracy after Pretesting with an Application to the Stock Market

D.L. Danilov, J.R. Magnus

Research output: Working paperDiscussion paperOther research output

255 Downloads (Pure)

Abstract

In econometrics, as a rule, the same data set is used to select the model and, conditional on the selected model, to forecast.However, one typically reports the properties of the (conditional) forecast, ignoring the fact that its properties are affected by the model selection (pretesting).This is wrong, and in this paper we show that the error can be very substantial.We obtain explicit expressions for this error.To illustrate the theory we consider the regression approach of Pesaran and Timmermann (1994) to stock market forecasting, and show that their proposed recursive predictions are much less robust than naive econometrics might suggest.
Original languageEnglish
Place of PublicationTilburg
PublisherEconometrics
Number of pages28
Volume2002-76
Publication statusPublished - 2002

Publication series

NameCentER Discussion Paper
Volume2002-76

Fingerprint

Econometrics
Stock market
Forecast accuracy
Prediction
Model selection

Keywords

  • forecasting
  • stock markets
  • return on investment

Cite this

Danilov, D. L., & Magnus, J. R. (2002). Forecast Accuracy after Pretesting with an Application to the Stock Market. (CentER Discussion Paper; Vol. 2002-76). Tilburg: Econometrics.
Danilov, D.L. ; Magnus, J.R. / Forecast Accuracy after Pretesting with an Application to the Stock Market. Tilburg : Econometrics, 2002. (CentER Discussion Paper).
@techreport{cb9b9b6340a94035924ed1fc04a994e1,
title = "Forecast Accuracy after Pretesting with an Application to the Stock Market",
abstract = "In econometrics, as a rule, the same data set is used to select the model and, conditional on the selected model, to forecast.However, one typically reports the properties of the (conditional) forecast, ignoring the fact that its properties are affected by the model selection (pretesting).This is wrong, and in this paper we show that the error can be very substantial.We obtain explicit expressions for this error.To illustrate the theory we consider the regression approach of Pesaran and Timmermann (1994) to stock market forecasting, and show that their proposed recursive predictions are much less robust than naive econometrics might suggest.",
keywords = "forecasting, stock markets, return on investment",
author = "D.L. Danilov and J.R. Magnus",
note = "Pagination: 28",
year = "2002",
language = "English",
volume = "2002-76",
series = "CentER Discussion Paper",
publisher = "Econometrics",
type = "WorkingPaper",
institution = "Econometrics",

}

Danilov, DL & Magnus, JR 2002 'Forecast Accuracy after Pretesting with an Application to the Stock Market' CentER Discussion Paper, vol. 2002-76, Econometrics, Tilburg.

Forecast Accuracy after Pretesting with an Application to the Stock Market. / Danilov, D.L.; Magnus, J.R.

Tilburg : Econometrics, 2002. (CentER Discussion Paper; Vol. 2002-76).

Research output: Working paperDiscussion paperOther research output

TY - UNPB

T1 - Forecast Accuracy after Pretesting with an Application to the Stock Market

AU - Danilov, D.L.

AU - Magnus, J.R.

N1 - Pagination: 28

PY - 2002

Y1 - 2002

N2 - In econometrics, as a rule, the same data set is used to select the model and, conditional on the selected model, to forecast.However, one typically reports the properties of the (conditional) forecast, ignoring the fact that its properties are affected by the model selection (pretesting).This is wrong, and in this paper we show that the error can be very substantial.We obtain explicit expressions for this error.To illustrate the theory we consider the regression approach of Pesaran and Timmermann (1994) to stock market forecasting, and show that their proposed recursive predictions are much less robust than naive econometrics might suggest.

AB - In econometrics, as a rule, the same data set is used to select the model and, conditional on the selected model, to forecast.However, one typically reports the properties of the (conditional) forecast, ignoring the fact that its properties are affected by the model selection (pretesting).This is wrong, and in this paper we show that the error can be very substantial.We obtain explicit expressions for this error.To illustrate the theory we consider the regression approach of Pesaran and Timmermann (1994) to stock market forecasting, and show that their proposed recursive predictions are much less robust than naive econometrics might suggest.

KW - forecasting

KW - stock markets

KW - return on investment

M3 - Discussion paper

VL - 2002-76

T3 - CentER Discussion Paper

BT - Forecast Accuracy after Pretesting with an Application to the Stock Market

PB - Econometrics

CY - Tilburg

ER -

Danilov DL, Magnus JR. Forecast Accuracy after Pretesting with an Application to the Stock Market. Tilburg: Econometrics. 2002. (CentER Discussion Paper).