Essays on causality over time /

This study offers three essays on causality over time. The first essay discusses the concept of causality and prediction. It explains two different approaches for identifying causal structure from observational data -- a regression based search procedure and a graphical procedure. The latter conne...

Full description

Bibliographic Details
Main Author: Guven, Derya
Format: Thesis Book
Language:English
Published: [Place of publication not identified] : [publisher not identified] ; 1996.
Subjects:
Online Access:http://proxy.library.tamu.edu/login?url=http://proquest.umi.com/pqdweb?did=743156901&sid=1&Fmt=2&clientId=2945&RQT=309&VName=PQD
Description
Summary:This study offers three essays on causality over time. The first essay discusses the concept of causality and prediction. It explains two different approaches for identifying causal structure from observational data -- a regression based search procedure and a graphical procedure. The latter connects probability with causality. Monte Carlo simulation is used to demonstrate the difference between these models. The second essay focuses these alternative procedures on observational data from the United States on exchange rates, corn price, corn export sales and corn export shipments. Out-of-sample forecasts from models identified using the alternatives are compared. Forecasts based on the graphical procedures can compete favorably with those based on a regression search procedure. Results may also be attributable to differences in two methods. The third essay considers the observational data on lumber price, lumber production, housing starts and disposable personal income. Again both graphical procedures and regression search are used in formulating models. There is a positive correlation between lumber price and production, but no direct causal relationship in contemporaneous time. Disposable income and housing starts affect price and quantities. Applications to two markets are presented because the differing characteristics of the data yield better insight into the strengths and weaknesses of the methods.
Item Description:Vita.
"Major Subject: Agricultural Economics".
Physical Description:xv, 164 leaves : illustrations ; 28 cm.
Issued also on microfiche from University Microfilms Inc.
Bibliography:Includes bibliographical references.