Enter a symbol:
For Toronto: TSX:TD

You are here :  Home > Articles > Page 3 > Persistent Winners Among Exchange Traded Funds

   This is a translation of the former paper published by André Gosselin on the OrientationFinance.com web site on June 27 2005 ( Read the original paper in French here).


   The Spectrum researches.

   The researches carried out by the mutual fund company Spectrum have enabled the validation of its "winner persistence" strategy on historical stock exchange data. As far as the industrial sectors are concerned, they used the TSE 300 monthly yields and the yields of its 14 sectorial indexes. The tested model consisted in comparing the return of a portfolio holding the 7 " winner" sectors of the TSE (those with the best momentum) with the return of a portfolio holding the 7 "losers" sectors known (those with worse the momentum). The winner portfolio was holding the 7 sectorial indexes which had recorded the best returns for the last 12 months, while the loser portfolio was holding the 7 sectorial indices which had known the worst returns for the last 12 months.

   Surprisingly, these winner and loser portfolios were not hold during 6 months or a year before being modified. The Spectrum strategists had preferred to adjust the portfolios once a month, with respect to the performance of each index over the last 12 trailing months. The results are all the same interesting, more especially as they refer to nearly 40 years of stock exchange history, from 1963 to 2000. During all this period, the 7 winner sectors portfolio has recorded an average annual yield of 15,1 %, against 8,5 % for the 7 loser sectors portfolio and 10,9 % for the TSE 300.

   As I mentioned above, the Spectrum Tactonics funds rely on the winner persistence at the sectorial level, at the national stock exchange markets level and at the management styles level. In order to have an overview of the historical yield of such an integrated strategy, the Spectrum strategists have used 18 American indexes related to the industrial sectors and the investment styles, and 6 indexes representing the countries. The winner and loser portfolios consisted of 12 indexes each, distributed according to their performance over the 12 more recent months and subjected to monthly adjustments. Because of the limits imposed by the availability of the stock exchange data bases, the validation could be made only over a very shortened period, going from 1994 to 2000.

   The portfolio holding the 12 winner indexes had generated, over this period, an average annual yield of 17,2 %, against 10,5 % for the portfolio holding the 12 loser indexes and 14,3 % for the MSCI World index. Although there is no perfect benchmark index for this type of portfolio, the Spectrum strategists had chosen the objective to exceed the MSCI World index.

   The philosophy of the Spectrum Tactonics funds does not only rely on the winner persistence of the indexes. It uses a rather nebulous risk measurement which generates a sale signal when "the yield of a market or a sector exceeds the expected yield in the long term and this, using two standard deviations". By using this indicator, it would be possible to add 1 to 2 percents to the annualized yield of the portfolio.

   The Spectrum validation study is extremely informative, in particular because it confirms the winner persistence within the sectorial indexes of the Toronto Stock Exchange. The phenomenon of sectorial momentum that American academics have recently discovered in the United States also seems to exist in Canada. When we will have access to enough ETF representing the sectorial sub-indexes of the Canadian economy (for the moment, they are only 2 or 3 of them), an investment strategy based on the sectorial winner persistence could be very attractive and lucrative.

   The product offered by Spectrum is certainly interesting and innovative. However, nothing prevents an investor from applying, by his own means, a winner persistence strategy. It is then enough to buy directly ETF rather than to invest in a fund of funds. As it is demonstrated by the university research, it is sufficient to buy the 4 or 5 best performer ETF of the last 12 months to have a gaining strategy, and to only make the necessary adjustments once a year.

   André Gosselin

Copyright ©2004-2023 Agnosoft