Using non-parametric search algorithms to forecast daily excess stock returns

Nathan Lael Joseph; David S. Brée; Efstathios Kalyvas

doi:10.1016/S0731-9053(04)19004-X

Using non-parametric search algorithms to forecast daily excess stock returns

Nathan Lael Joseph, David S. Brée, Efstathios Kalyvas

Research output: Chapter in Book/Report/Conference proceeding › Chapter › peer-review

Abstract

Are the learning procedures of genetic algorithms (GAs) able to generate optimal architectures for artificial neural networks (ANNs) in high frequency data? In this experimental study, GAs are used to identify the best architecture for ANNs. Additional learning is undertaken by the ANNs to forecast daily excess stock returns. No ANN architectures were able to outperform a random walk, despite the finding of non-linearity in the excess returns. This failure is attributed to the absence of suitable ANN structures and further implies that researchers need to be cautious when making inferences from ANN results that use high frequency data. © 2004 Elsevier Ltd. All rights reserved.

Original language	English
Title of host publication	Advances in Econometrics
Publisher	JAI Press
Pages	93-125
Number of pages	33
Volume	19
ISBN (Electronic)	978-1-84950-303-7
ISBN (Print)	0762311509, 978-0-76231-150-7
DOIs	https://doi.org/10.1016/S0731-9053(04)19004-X
Publication status	Published - 1 Jan 2004

Publication series

Name	Advances in Econometrics
Volume	19

Access to Document

10.1016/S0731-9053(04)19004-X

Cite this

@inbook{c1509c0e6c004442a3528dcf93d85d47,

title = "Using non-parametric search algorithms to forecast daily excess stock returns",

abstract = "Are the learning procedures of genetic algorithms (GAs) able to generate optimal architectures for artificial neural networks (ANNs) in high frequency data? In this experimental study, GAs are used to identify the best architecture for ANNs. Additional learning is undertaken by the ANNs to forecast daily excess stock returns. No ANN architectures were able to outperform a random walk, despite the finding of non-linearity in the excess returns. This failure is attributed to the absence of suitable ANN structures and further implies that researchers need to be cautious when making inferences from ANN results that use high frequency data. {\textcopyright} 2004 Elsevier Ltd. All rights reserved.",

author = "Joseph, {Nathan Lael} and Br{\'e}e, {David S.} and Efstathios Kalyvas",

year = "2004",

month = jan,

day = "1",

doi = "10.1016/S0731-9053(04)19004-X",

language = "English",

isbn = "0762311509",

volume = "19",

series = "Advances in Econometrics",

publisher = "JAI Press",

pages = "93--125",

booktitle = "Advances in Econometrics",

address = "United States",

}

TY - CHAP

T1 - Using non-parametric search algorithms to forecast daily excess stock returns

AU - Joseph, Nathan Lael

AU - Brée, David S.

AU - Kalyvas, Efstathios

PY - 2004/1/1

Y1 - 2004/1/1

N2 - Are the learning procedures of genetic algorithms (GAs) able to generate optimal architectures for artificial neural networks (ANNs) in high frequency data? In this experimental study, GAs are used to identify the best architecture for ANNs. Additional learning is undertaken by the ANNs to forecast daily excess stock returns. No ANN architectures were able to outperform a random walk, despite the finding of non-linearity in the excess returns. This failure is attributed to the absence of suitable ANN structures and further implies that researchers need to be cautious when making inferences from ANN results that use high frequency data. © 2004 Elsevier Ltd. All rights reserved.

AB - Are the learning procedures of genetic algorithms (GAs) able to generate optimal architectures for artificial neural networks (ANNs) in high frequency data? In this experimental study, GAs are used to identify the best architecture for ANNs. Additional learning is undertaken by the ANNs to forecast daily excess stock returns. No ANN architectures were able to outperform a random walk, despite the finding of non-linearity in the excess returns. This failure is attributed to the absence of suitable ANN structures and further implies that researchers need to be cautious when making inferences from ANN results that use high frequency data. © 2004 Elsevier Ltd. All rights reserved.

UR - https://www.mendeley.com/catalogue/6e2fedcc-a8f4-3f2e-951b-fec98d871df1/

U2 - 10.1016/S0731-9053(04)19004-X

DO - 10.1016/S0731-9053(04)19004-X

M3 - Chapter

SN - 0762311509

SN - 978-0-76231-150-7

VL - 19

T3 - Advances in Econometrics

SP - 93

EP - 125

BT - Advances in Econometrics

PB - JAI Press

ER -

Using non-parametric search algorithms to forecast daily excess stock returns

Abstract

Publication series

Access to Document

Other files and links

Fingerprint

Cite this