Return volatility and the pricing of equities at the Nairobi Securities Exchange
Date
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
The study is an empirical test of asset pricing theory, which postulates that volatility is priced if
a positive relationship exists between asset returns and their volatility. The asset under study
is the Nairobi Securities Exchange 20-Share Index (NSE-20), which is assumed to be a welldiversified
portfolio. Data from daily values of the NSE-20 were collected over a 15 year period
from January 1999 to December 2013, and from these data, monthly returns were computed.
Two models of the Auto-Regressive Conditional Heteroscedasticity-in-Mean (ARCH-M) family
allowed the study to specify volatility conditionally. The study finds that monthly NSE-20
returns are not normally distributed and exhibit volatility clustering. A positive and significant
relationship between risk and returns is found; indicating that volatility is priced. The study
hopes to be useful to portfolio managers when carrying out a forward-looking valuation of a
well-diversified portfolio of Kenyan stocks as well as other similar stocks based on market
characteristics.
Description
Paper presented at the 11th Africa Finance Journal Conference, Durban, South Africa.
The study is an empirical test of asset pricing theory, which postulates that volatility is priced if a positive relationship exists between asset returns and their volatility. The asset under study is the Nairobi Securities Exchange 20-Share Index (NSE-20), which is assumed to be a welldiversified portfolio. Data from daily values of the NSE-20 were collected over a 15 year period from January 1999 to December 2013, and from these data, monthly returns were computed. Two models of the Auto-Regressive Conditional Heteroscedasticity-in-Mean (ARCH-M) family allowed the study to specify volatility conditionally. The study finds that monthly NSE-20 returns are not normally distributed and exhibit volatility clustering. A positive and significant relationship between risk and returns is found; indicating that volatility is priced. The study hopes to be useful to portfolio managers when carrying out a forward-looking valuation of a well-diversified portfolio of Kenyan stocks as well as other similar stocks based on market characteristics.
The study is an empirical test of asset pricing theory, which postulates that volatility is priced if a positive relationship exists between asset returns and their volatility. The asset under study is the Nairobi Securities Exchange 20-Share Index (NSE-20), which is assumed to be a welldiversified portfolio. Data from daily values of the NSE-20 were collected over a 15 year period from January 1999 to December 2013, and from these data, monthly returns were computed. Two models of the Auto-Regressive Conditional Heteroscedasticity-in-Mean (ARCH-M) family allowed the study to specify volatility conditionally. The study finds that monthly NSE-20 returns are not normally distributed and exhibit volatility clustering. A positive and significant relationship between risk and returns is found; indicating that volatility is priced. The study hopes to be useful to portfolio managers when carrying out a forward-looking valuation of a well-diversified portfolio of Kenyan stocks as well as other similar stocks based on market characteristics.
Keywords
Nairobi Securities Exchange, markets, pricing, volatility