Determining the accuracy of the Nairobi stock exchange 20-share index

Abstract

The study set out to determine the accuracy of the Nairobi Stock Exchange (NSE) 20- .J Share index. The need for the study emanated from the apparent weakness of the NSE 20- Share index to effectively report on the market performance as it is calculated using unadjusted share prices and a geometric averaging method that has been statistically proven to have a downward bias in the long run. In addition, there is need for an alternative performance measure to give a balanced view value movements in the NSE. The derived indices were four in number: a Composite All-Share index, a 20-Share index (Laspeyres method), a 20- Share index (Paasche method) and a 20-Share index (Fisher's ideal method). They all used 1996 week 14 as the base period and the base value was set at 3007.39,the value of the NSE 20-Share index in the same period. This figure was selected to provide for uniformity and enable comparison among the indices. The share prices were adjusted for bonus issues throughout the period under study and the number of shares in issue were used as the weights for all the indices derived. During the four years under study, the N-SE 20-Share index was found to have understated price rises by 1208%, rendering the index inaccurate and unsuitable for measuring:lone-term rice movements. A more accurate index calculated using the Fisher's Ideal formula was recommended for the Nairobi Stock Exchange as it proved to able to effectively capture and report on market performance.