sorry for duplicate...this might interest some....it was a study done by Yang over at astrikos.
Don't think they would mind me posting it as it will probably drive a few page views over there.
Speaking of moving averages, we've been conducting some studies recently
that I think you might find interesting. Take the simple strategy of
using a moving average's trend to determine buy and sell signals in the
S&P500. If the moving average is trending higher, it's a buy signal,
while a moving average trending lower is a sell signal. Using this
simple strategy, we were interested in discovering which moving averages
worked best when it comes to the S&P500. Common technical analysis
usually pushes the 20 day, 50 day or 200 day as some of the 'best'
moving averages. Our work says otherwise. In fact, if I were to ask you
which moving average between 1 and 200 performed best using the simple
buy and sell rules outlined above, which would you guess? I bet you'd
never guess '1'. A 1-day moving average, of course, really isn't a
moving average at all - it's simply the day's close. By going long the
S&P500 whenever the 1-day average (today's close) is higher than the
previous day's 1-day average (yesterday's close) and short the S&P500
whenever the 1-day average (today's close) is lower than the previous
day's 1-day average (yesterday's close), an investor would have amassed
a staggering 70,000% profit (yes, 70,000%) since 1970. That outperforms
any other technical strategy that we've ever come across, including the
S&P Institutional Money Flow. Of course, a manic trading strategy like
this one isn't for everyone - after all, there were 3,800 closed trades
since 1970, meaning you'd be switching from long to short and vice versa
roughly every two days. Still, profits of that magnitude would more than
make up for transaction costs. Something to consider the next time
someone touts the effectiveness of the 10, 20 or 50-day moving averages.
What's really important, this study illustrates, is what's happening
right now. The table below lists the top ten moving averages sorted by
their overall profit trading the S&P500...
1-day average: +70,148%
2-day average: +14,834%
114-day average: +1,305%
123-day average: +1,266%
3-day average: +1,265%
113-day average: +1,117%
111-day average: +1,088%
117-day average: +1,079%
125-day average: +1,069%
112-day average: +1,057%
Interesting to note from the table that five of top ten best performing
moving averages were in the 110-120 day range, or roughly a six-month
average on the daily charts. If you're looking for a longer-term moving
average to incorporate into your work, one of these might fit the bill
nicely. And just in case you're interested in how some of the more
popular averages fared using the same buy and sell rules outlined above,
here's a quick look...
10-day average: +51%
20-day average: +195%
50-day average: +103%
200-day average: +252%
(remember, these figures are over a 32-year period)
From Yang.. technical trader...