Momentum Investing In Mutual Funds

RESULTS OF THE SYSTEM
Based on these emails, I decided to run the strategy for one year using actual money, and publish my results on a web page for clients to see and observe. In the first month, January 2001, the model portfolio of five Canadian mutual funds showed a loss of 3.2%. From February, however, it started rising slowly, to finish the year with an 8.15% profit.

This may not seem like a huge profit, but consider that the North American market in 2001 was moving into a recession, depressed even further by the 9/11 attacks. In the same year, the Dow Jones lost approximately 6%, and the Toronto Stock Exchange dropped by a similar percentage. Now consider that the portfolio moved in the opposite direction; with that in mind, 8.15% is indeed a remarkable profit.

I became so impressed with the strategy that on September 1, 2001, I decided to start a broader fund portfolio of mutual funds, a sector fund portfolio of Index funds, and an i-Unit portfolio of exchange-traded funds (ETF's) found on the Toronto Stock Exchange (TSE). The broad portfolio consisted of 20 funds two funds each out of 10 various families that I watched. The sector portfolio consisted of one fund family, CI mutual funds, where the funds only invest in a sector of the market, almost like an index fund. The i-Unit portfolio was made up of exchange-traded funds (ETF's) that are similar to i-Shares on the US market. The index units are traded on the Toronto stock market exactly like stocks.

TWEAKING THE SYSTEM
In the months leading to September, I changed the strategy for the bear market. Having a stop-loss from the highest high over the past two years was no longer necessary. In my original tests in 1999, I looked at two years of history where, during those two years, three-quarters were a bull market and the rest a bear market.

In addition, I found I needed a trigger based on the market trend. If a fund/i-unit gave me a sell signal, via a stop-loss or a move below the 28-week moving average, I would sell the fund/i-Unit. However, if the trigger was negative, I would move into the money market rather than into the next best-performing fund. When the September 2001 tragedy occurred in the US, my model portfolio of five funds was almost 62% in the money market; one week after September 11, it moved into positive territory allowing me to move from the money market into the top-performing funds.

I also changed the stop-loss technique. Instead of using a stop-loss of 12% from the highest high reached over the two years, I used a 12% stop-loss from the entry price with a rising stop. Remember, too, that I started three other portfolios on September 1. With 9/11, they all gave sell signals on most of the funds that made up the portfolios, with the monies going into the money market. However, a week later, they were back into funds as the trigger became positive. This pleased me, because the only flaw I could now see in the strategy was when to move new client money into the portfolio. I could buy today as an entry into a portfolio and receive a sell signal tomorrow. In spite of this flaw, I found that once the strategy became operative, it was a simple automatic strategy.

Most profits came from the period after the 9/11 crisis, as suggested by the broad model portfolio. The sector fund portfolio was most disappointing, and I scrapped it in March 2002. I also wanted to start a portfolio of Fidelity Focus funds, but the range offered in Canada is not as great as that offered in the US.

The winner out of all the portfolios, the one that gave me the greatest profits, was the i-Unit portfolio. Its performance excelled simply because it gave a buy signal on the S&P/TSE Canadian Gold Index (XIT) soon after 9/11.

More recently, for January 1, 2005, to August 18, 2005, the portfolios show the profits you see in Figure 1 (i-Unit Portfolio).

Figure 1: i-Unit portfolio. Profits for January 1, 2005, to August 18, 2005.

All the portfolios are on an Excel spreadsheet. Prices are updated every Saturday morning. Figures 2 and 3 illustrate two more examples of the many portfolios I monitor. Figure 2 is the Growth Portfolio, not doing as well for 2005 as I had hoped. Figure 3 is the High Income Fund portfolio, currently the name of the game in Canadian Mutual Funds.

Figure 2: Growth Portfolio.

Figure 3: High income fund portfolio.

The columns of the Excel spreadsheet are as follows:

Security name.
Fund code
Close above or below a 28 week moving average(calculated using Technifilter Plus)
Stop-loss(calculated as 12% from entry price) if((P-J)/P) > 0.12,"Sell","Hold")
Units purchased
Price paid
Current price
Cost + commission H * I
Market valueH * J
Gain/lossL - K
Gain/loss %M / K
% of portfolioK / Total K

New high =IF( New High>High reached,New," ") If(P>R,"New"," ")
Stop-loss ==IF(((New high-current price)/New high)>0.12,"Sell","hold")

In other words, after calculating the new high, if today's closing price is less than the new high of -12%, then sell.

The last column, "New high," warns me that a new high has been reached. The formula for that column is:

If current price > high reached then show "New ". if(J>R,"New"," ")

I alter the "High reached" column manually every time a New high is shown.

Jack Singer, PhC, has been a technical analyst since 1969 and has worked as a futures and options trader, a research technician specializing in gold and gold shares, and editor of the South African tipping sheet Temkin and Moon, before emigrating to Canada, where he qualified as an investment advisor. He can be reached at jasin@shaw.ca.

RELATED READING
Singer, Jack [2000]. "Enhancing Fund Switching," Technical Analysis of Stocks & Commodities, Volume 18: October.

Current and past articles from Working Money, The Investors' Magazine, can be found at Working-Money.com.