Information, data dimension and factor structure

Jan P. A. M. Jacobs; Pieter W. Otter; Ard H. J. den Reijer

doi:10.1016/j.jmva.2011.11.003

Information, data dimension and factor structure

Jan P. A. M. Jacobs^*, Pieter W. Otter, Ard H. J. den Reijer

^*Corresponding author for this work

Economics, Econometrics and Finance

Research output: Contribution to journal › Article › Academic › peer-review

1 Citation (Scopus)

183 Downloads (Pure)

Abstract

This paper employs concepts from information theory for choosing the dimension of a data set. We propose a relative information measure connected to Kullback-Leibler numbers. By ordering the series of the data set according to the measure, we are able to obtain a subset of a data set that is most informative. The method can be used as a first step in the construction of a dynamic factor model or a leading index, as illustrated with a Monte Carlo study and with the US macroeconomic data set of Stock and Watson [20]. (C) 2011 Elsevier Inc. All rights reserved.

Original language	English
Pages (from-to)	80-91
Number of pages	12
Journal	Journal of Multivariate Analysis
Volume	106
DOIs	https://doi.org/10.1016/j.jmva.2011.11.003
Publication status	Published - Apr-2012

Keywords

Kullback-Leibler numbers
Information
Factor structure
Data set dimension
Dynamic factor models
Leading index
DYNAMIC-FACTOR MODEL
NUMBER
PREDICTORS
SELECTION

Access to Document

10.1016/j.jmva.2011.11.003

Information, data dimension and factor structureFinal publisher's version, 278 KBLicence: Elsevier

Handle.net

Persistent link

Cite this

@article{fb79b16119334731b09676dba9927211,

title = "Information, data dimension and factor structure",

abstract = "This paper employs concepts from information theory for choosing the dimension of a data set. We propose a relative information measure connected to Kullback-Leibler numbers. By ordering the series of the data set according to the measure, we are able to obtain a subset of a data set that is most informative. The method can be used as a first step in the construction of a dynamic factor model or a leading index, as illustrated with a Monte Carlo study and with the US macroeconomic data set of Stock and Watson [20]. (C) 2011 Elsevier Inc. All rights reserved.",

keywords = "Kullback-Leibler numbers, Information, Factor structure, Data set dimension, Dynamic factor models, Leading index, DYNAMIC-FACTOR MODEL, NUMBER, PREDICTORS, SELECTION",

author = "Jacobs, {Jan P. A. M.} and Otter, {Pieter W.} and {den Reijer}, {Ard H. J.}",

year = "2012",

month = apr,

doi = "10.1016/j.jmva.2011.11.003",

language = "English",

volume = "106",

pages = "80--91",

journal = "Journal of Multivariate Analysis",

}

TY - JOUR

T1 - Information, data dimension and factor structure

AU - Jacobs, Jan P. A. M.

AU - Otter, Pieter W.

AU - den Reijer, Ard H. J.

PY - 2012/4

Y1 - 2012/4

N2 - This paper employs concepts from information theory for choosing the dimension of a data set. We propose a relative information measure connected to Kullback-Leibler numbers. By ordering the series of the data set according to the measure, we are able to obtain a subset of a data set that is most informative. The method can be used as a first step in the construction of a dynamic factor model or a leading index, as illustrated with a Monte Carlo study and with the US macroeconomic data set of Stock and Watson [20]. (C) 2011 Elsevier Inc. All rights reserved.

AB - This paper employs concepts from information theory for choosing the dimension of a data set. We propose a relative information measure connected to Kullback-Leibler numbers. By ordering the series of the data set according to the measure, we are able to obtain a subset of a data set that is most informative. The method can be used as a first step in the construction of a dynamic factor model or a leading index, as illustrated with a Monte Carlo study and with the US macroeconomic data set of Stock and Watson [20]. (C) 2011 Elsevier Inc. All rights reserved.

KW - Kullback-Leibler numbers

KW - Information

KW - Factor structure

KW - Data set dimension

KW - Dynamic factor models

KW - Leading index

KW - DYNAMIC-FACTOR MODEL

KW - NUMBER

KW - PREDICTORS

KW - SELECTION

U2 - 10.1016/j.jmva.2011.11.003

DO - 10.1016/j.jmva.2011.11.003

M3 - Article

VL - 106

SP - 80

EP - 91

JO - Journal of Multivariate Analysis

JF - Journal of Multivariate Analysis

ER -

Information, data dimension and factor structure

Abstract

Keywords

Access to Document

Handle.net

Fingerprint

Cite this