Jack, my switching method is short-term. It's basically a "let your winners run and cut your losses short" system. The conservative methods are in cash about 80% of the time, only entering the stock market on the days that are rated as extremely likely to have a good gain. The moderate methods are in cash maybe 50% of the time.
The big benefit of the methods is that you substantially avoid the dips. For example, in the Sept-Oct 1999 dip of 13% in the NYSE, my actual losses were about 5%, yet I've gained close to 30% in the Nov-Jan run-up. So, by avoiding the dips, your long-term return is significantly improved. It's different from the AIM methods, in that my methods don't know if the market is high or low, they just sense changes in direction. But it's similar in that you buy low and sell high, but my lows and highs are short-term. For example, I was in cash early last week, then bought FFIDX on Thursday, in time for the nice Friday and Monday gains, then sold on Tuesday because the models said today was a risky day. Tomorrow's a risky day too according to the models, so I'm still in cash.
I normally buy and hold from November through January (see my "January Effect" study), but this year I exited because the NASDAQ runup was so huge and I had a big profit already from Nov-Dec, so now I'm back to my short-term methods.
My methods are "skittish" -- they exit the market at the first sign of danger. So, I don't mind skating close to the peaks. However, if volatility gets too high, the methods all go into cash. A choppy market is a killer for short-term trading, so I don't fool around when things get crazy.
Percent of winners and losers? Well, about 50-50, actually. The losers are often sold after one day. The winners are usually held onto for many days. As I said, cut the losses short, let the winners run.
Now--that points out the problem: transactions. This method works great if you have very low or no transaction cost. This is my situation (I'm trading a Fidelity retirement account using primarily their no-load funds). I've calculated that if you traded in the open market using a broker to trade QQQ or SPY, you'll suffer a 1% performance degradation if you're trading $80,000, 2% if you're trading $40,000, 4% if you're trading $20,000. If you're only trading $10,000, you'd have to use my one aggressive strategy (Model 17) to overcome the cost of transactions (I'm assuming an $8 per trade transaction fee).
You're in an index fund. Is it no-load? Do they charge you if you trade in and out? If not, my methods would work with any portfolio value.
I'm starting a free subscription service. The subscribers will be notified in advance when I am about to make a trade. Also, I'm going to start updating the web site several times during the day, so people will know what the models think of the day's action, and can trade with me.
You can make 30% long-term very safely with these methods. Look at my report "Stock Market Modeling Techniques and Potential Applications". My Model #9 gained 23% annually in a stock market that gained 8% annually over the past 30 years (and it didn't lose money any year, including the wicked '73-74 50% bear market). More aggressive models gained more. It all depends on what draw-down level you're willing to tolerate. My tolerance level is very low (2 kids closing in on college age), so I tend to take the conservative route. But I intend to present multiple levels of risk on the new version of the site.
Thanks for your interest!
Kevin Farnham |
