Connectionist-based rules describing the pass-through of individual goods prices into trend inflation in the United States

Vincent A. Schmidt; Jane M. Binner; Richard G. Anderson

Connectionist-based rules describing the pass-through of individual goods prices into trend inflation in the United States

Vincent A. Schmidt^*, Jane M. Binner, Richard G. Anderson

^*Corresponding author for this work

Economics, Finance and Entrepreneurship

Research output: Chapter in Book/Published conference output › Conference publication

Abstract

This paper demonstrates a mechanism whereby rules can be extracted from a feedforward neural network trained to characterize the inflation "pass-through" problem in American monetary policy, defined as the relationship between changes in the growth rate(s) of individual commodities and the economy-wide rate of growth of consumer prices. Monthly price data are encoded and used to train a group of candidate connectionist architectures. One candidate is selected for rule extraction, using a custom decompositional extraction algorithm that generates rules in human-readable and machine-executable form. Rule and network accuracy are compared, and comments are made on the relationships expressed within the discovered rules. The types of discovered relationships could be used to guide monetary policy decisions.

Original language	English
Title of host publication	Proceedings of the 2010 International Conference on Artificial Intelligence, ICAI 2010
Editors	Hamid R. Arabnia, David de la Fuente, Elena B. Kozerenko, et al
Publisher	CSREA
Pages	318-324
Number of pages	7
ISBN (Print)	978-1-6013-2148-0
Publication status	Published - 1 Dec 2010
Event	2010 International Conference on Artificial Intelligence - Las Vegas, NV, United States Duration: 12 Jul 2010 → 15 Jul 2010

Conference

Conference	2010 International Conference on Artificial Intelligence
Abbreviated title	IC-AI 2010
Country/Territory	United States
City	Las Vegas, NV
Period	12/07/10 → 15/07/10

Keywords

consumer prices
data mining
inflation
neural network
rule generation

Cite this

Schmidt, Vincent A. ; Binner, Jane M. ; Anderson, Richard G. / Connectionist-based rules describing the pass-through of individual goods prices into trend inflation in the United States. Proceedings of the 2010 International Conference on Artificial Intelligence, ICAI 2010. editor / Hamid R. Arabnia ; David de la Fuente ; Elena B. Kozerenko ; et al. CSREA, 2010. pp. 318-324

@inproceedings{402dbf87241c49ee91d4699f01d36488,

title = "Connectionist-based rules describing the pass-through of individual goods prices into trend inflation in the United States",

abstract = "This paper demonstrates a mechanism whereby rules can be extracted from a feedforward neural network trained to characterize the inflation {"}pass-through{"} problem in American monetary policy, defined as the relationship between changes in the growth rate(s) of individual commodities and the economy-wide rate of growth of consumer prices. Monthly price data are encoded and used to train a group of candidate connectionist architectures. One candidate is selected for rule extraction, using a custom decompositional extraction algorithm that generates rules in human-readable and machine-executable form. Rule and network accuracy are compared, and comments are made on the relationships expressed within the discovered rules. The types of discovered relationships could be used to guide monetary policy decisions.",

keywords = "consumer prices, data mining, inflation, neural network, rule generation",

author = "Schmidt, {Vincent A.} and Binner, {Jane M.} and Anderson, {Richard G.}",

year = "2010",

month = dec,

day = "1",

language = "English",

isbn = "978-1-6013-2148-0",

pages = "318--324",

editor = "Arabnia, {Hamid R.} and {de la Fuente}, David and Kozerenko, {Elena B.} and {et al}",

booktitle = "Proceedings of the 2010 International Conference on Artificial Intelligence, ICAI 2010",

publisher = "CSREA",

note = "2010 International Conference on Artificial Intelligence , IC-AI 2010 ; Conference date: 12-07-2010 Through 15-07-2010",

}

Schmidt, VA, Binner, JM & Anderson, RG 2010, Connectionist-based rules describing the pass-through of individual goods prices into trend inflation in the United States. in HR Arabnia, D de la Fuente, EB Kozerenko & et al (eds), Proceedings of the 2010 International Conference on Artificial Intelligence, ICAI 2010. CSREA, pp. 318-324, 2010 International Conference on Artificial Intelligence , Las Vegas, NV, United States, 12/07/10.

Connectionist-based rules describing the pass-through of individual goods prices into trend inflation in the United States. / Schmidt, Vincent A.; Binner, Jane M.; Anderson, Richard G.
Proceedings of the 2010 International Conference on Artificial Intelligence, ICAI 2010. ed. / Hamid R. Arabnia; David de la Fuente; Elena B. Kozerenko; et al. CSREA, 2010. p. 318-324.

Research output: Chapter in Book/Published conference output › Conference publication

TY - GEN

T1 - Connectionist-based rules describing the pass-through of individual goods prices into trend inflation in the United States

AU - Schmidt, Vincent A.

AU - Binner, Jane M.

AU - Anderson, Richard G.

PY - 2010/12/1

Y1 - 2010/12/1

N2 - This paper demonstrates a mechanism whereby rules can be extracted from a feedforward neural network trained to characterize the inflation "pass-through" problem in American monetary policy, defined as the relationship between changes in the growth rate(s) of individual commodities and the economy-wide rate of growth of consumer prices. Monthly price data are encoded and used to train a group of candidate connectionist architectures. One candidate is selected for rule extraction, using a custom decompositional extraction algorithm that generates rules in human-readable and machine-executable form. Rule and network accuracy are compared, and comments are made on the relationships expressed within the discovered rules. The types of discovered relationships could be used to guide monetary policy decisions.

AB - This paper demonstrates a mechanism whereby rules can be extracted from a feedforward neural network trained to characterize the inflation "pass-through" problem in American monetary policy, defined as the relationship between changes in the growth rate(s) of individual commodities and the economy-wide rate of growth of consumer prices. Monthly price data are encoded and used to train a group of candidate connectionist architectures. One candidate is selected for rule extraction, using a custom decompositional extraction algorithm that generates rules in human-readable and machine-executable form. Rule and network accuracy are compared, and comments are made on the relationships expressed within the discovered rules. The types of discovered relationships could be used to guide monetary policy decisions.

KW - consumer prices

KW - data mining

KW - inflation

KW - neural network

KW - rule generation

UR - http://www.scopus.com/inward/record.url?scp=84866085866&partnerID=8YFLogxK

M3 - Conference publication

AN - SCOPUS:84866085866

SN - 978-1-6013-2148-0

SP - 318

EP - 324

BT - Proceedings of the 2010 International Conference on Artificial Intelligence, ICAI 2010

A2 - Arabnia, Hamid R.

A2 - de la Fuente, David

A2 - Kozerenko, Elena B.

A2 - et al,

PB - CSREA

T2 - 2010 International Conference on Artificial Intelligence

Y2 - 12 July 2010 through 15 July 2010

ER -

Connectionist-based rules describing the pass-through of individual goods prices into trend inflation in the United States

Abstract

Conference

Keywords

Other files and links

Fingerprint

Cite this