COMPARISON OF DIFFERENT VARIANCE RATIO TESTS UNDER HETEROSKEDASTICITY AND AUTOCORRELATION: AN APPLICATION TO CHEMICAL SECTOR DATA IN PAKISTAN STOCK MARKET

Received: 25.04.2021; Revised: 28.06.2021, Accepted: 05.07.2021, Published Online: 31.07.2021

Nuzhat Aftab

College of Statistical and Actuarial Sciences, University of the Punjab, Lahore, Pakistan HED Punjab, Lahore, Pakistan, m.nuzhat.aftab@gmail.com

Amna Riaz

University Of Gujrat, Gujrat Pakistan amna.riaz@uog.edu.pk

Muhammad Suhail

The University of Agriculture, Peshawar, Pakistan msuhail88@aup.edu.pk

Iqra Baber

College of Statistical and Actuarial Sciences,University of the Punjab, Lahore, Pakistan

Anum Zakir

Virtual University of Pakistan, Lahore, Pakistan anam.zakir@vu.edu.pk

Khadija Fatima

University Of Gujrat, Gujrat Pakistan khadija.fatima@uog.edu.pk

Ijlal Haider

Faculty of Management Sciences, The University of Agriculture Peshawar, Pakistan haidersijlal29@gmail.com

Abstract

In this article, we used different variance ratio tests to evaluate the performance of firms registered in the chemical sector of Pakistan stock market in the presence of heteroskedasticity and autocorrelation. The existing variance ratio tests are applied to the daily and weekly observed and corrected returns.   It is concluded that the series are more efficient for corrected daily and weekly series. Moreover, the results showed that the weighted variance ratio tests are more consistent under both heteroskedasticity and homoskedasticity.

Keywords: Heteroskedasticity: Autocorrelation:  Bootstrap