Published share tips : do they out-perform the JSE?

Voigt, Ivan (2001)

Study project (MBA) -- University of Stellenbosch, 2001.

University of Stellenbosch Business School


ENGLISH ABSTRACT: This study considers share tips published in a respected publication, and determines whether an investment strategy based on the recommendations of its journalists could allow investors to exceed the stock market average. Six journalists were selected, and the recommendations that they made over a 30-month period grouped into “buy” and “do not buy” recommendations. The change in price of the recommended shares was measured after periods of one week, one month, three months and six months after the date of publication and after inclusion of dividends paid during those periods, returns were calculated. The returns attained for each share was compared to the return on the JSE-Overall Index during that period, the difference between the two being the excess return of the share. The excess returns of the shares recommended by each journalist were used to calculate portfolio excess returns, on which tests of statistical significance carried out. The portfolio of one journalist showed statistically significant excess returns in all four periods under review. One other achieved a statistically significant excess return over 1 week. No other portfolios achieved significant excess returns over the market.

AFRIKAANSE OPSOMMING: In hierdie werkstuk word die aandeelwenke wat in ‘n gerespekteerde tydskrif gepubliseer is, ondersoek om vas te stel of ‘n beleggingsstrategie wat op die wenke van die joernaliste gebaseer is, die mark gemiddlede opbrengs kan klop. Ses joernaliste is gekies, en hul wenke oor ‘n periode van 30-maande is geklassifiseer in “koop” en “nie koop” wenke. Vir die “koop” wenke is die prys-verandering oor tydperke van een week, een maand, drie maande en ses maande gemeet. Opbrengste met insluiting van dividende is bereken. Die opbrengste is met die JE-algehele indeks se opbrengs vir elk van die periodes vergelyk, en die verskil is as bo-opbrengste gedefinieer. Die bo-opbrengste vir elke aandeel is gebruik om portfolio bo-opbrengste te bereken, weereens vir elk van die periodes. Hierdie bo-opbrengste is vir statistiese betekenisvolheid getoets. Die portfolio van een joernalis het statisties beteksnisvolle bo-opbrengste vir al vier periodes getoon. Die portfolio van een ander joernalis het statisties betekenisvolle bo-opbrengste vir ‘n hou-periode van een week getoon. Geen ander portfolios het bo-opbrengste getoon nie.

Please refer to this item in SUNScholar by using the following persistent URL:
This item appears in the following collections: