Structural interdependence and efficiency in Pakistan's financial markets /

The concept of cointegration has become a cornerstone in applied time series analysis to study the long-run equilibrium relationship among the series. The idea was originally put forth by Engle and Granger (1987). According to the Granger Representation Theorem, cointegrated series can be represent...

Full description

Bibliographic Details
Main Author: Khan, Asim Muzaffar, 1966-
Format: Thesis Book
Language:English
Published: [Place of publication not identified] : [publisher not identified] ; 2002.
Subjects:
Online Access:http://proxy.library.tamu.edu/login?url=http://proquest.umi.com/pqdweb?did=726444511&sid=1&Fmt=2&clientId=2945&RQT=309&VName=PQD
Description
Summary:The concept of cointegration has become a cornerstone in applied time series analysis to study the long-run equilibrium relationship among the series. The idea was originally put forth by Engle and Granger (1987). According to the Granger Representation Theorem, cointegrated series can be represented by an error correction model. By incorporating the error correction term one should improve upon the long-run forecasting ability of the model. However, some studies have concluded that it may not be always the case. The concept of cointegration has one important implication for the Efficient Market Hypothesis. The evidence of cointegration points to the presence of inefficiency in the market since it implies that one can better forecast by utilizing the underlying long-run relationship among the series. A number of studies have looked at the evidence of cointegration in stock market data and in the foreign exchange rate markets in developed countries, but none has studied the underdeveloped economies such as Pakistan. Using Johansen's (1992) test procedure, three stock markets and the foreign exchange rate markets for eight major trading partners of Pakistan are studied. The major findings of the study include the presence of one cointegrating vector in stock markets and four cointegrating vectors in exchange rates markets. The evidence of cointegration indicates market inefficiency in the financial sectors of Pakistan's economy. However, one should bear in mind that the models did not explicitly incorporate any transaction costs involved in the stock markets and the foreign exchange rate markets. The omission of these transaction costs may result in inefficiency in an otherwise efficient market. This warrants further research in this field. There is an important implication of cointegration from a diversification point of view. If some markets are in cointegration relationships, then any diversification across those markets would be a futile exercise because any negative shock in one market will be transmitted to the other market as well. The models' performance criteria like RMSE and Theil U suggest that there is a gain in terms of better forecasting from modeling the stock markets and foreign exchange rate markets in VECM framework.
Item Description:Vita.
"Major Subject: Agricultural Economics".
Physical Description:xi, 137 leaves : illustrations ; 28 cm.
Issued also on microfiche from University Microfilm Inc.
Bibliography:Includes bibliographical references (leaves 132-136).