Hi Don-
With the s/w stuff separated, I'll try to be a little more specific about you're usage. Keep in mind that this is only opinion and that there are a lot of ways to peel the onion.
re: 4,300+ symbol database... maintaining 4,300+ symbols, you're spending too much time on data maintenance. And if you aren't spending a lot of time maintaining the data, you aren't reviewing the daily data variance reports and correcting all the errors. That means that the reports you're generating are as filled with errors as there are errors in the variance report (not all variances are legitimate errors, but they have to be tracked down and confirmed). It's mandatory to have clean data and then run the reports on the clean data. Running the reports then correcting the data does no good and that includes the market signals. AIQ recommends a minimum volume (which I think is a little too low) and they did most of their testing using the S&P 500.
With 2,000+ symbols I was spending 1 to 2 hours per day correcting data errors and looking up symbol changes. Your numbers have to be higher than that. Once I went thru and got rid of every stock that didn't trade every day and stocks that were less than $2/sh. and more than $50 I was able to get it down to 1/2 hr per night.
Another way to approach it is to do a fundamentals search using any of the search facilities (Prosearch, Stock Investor, Value Line, etc.). You can also go to IBD and copy into your database all the stocks that have an earnings rank (ER) of greater than some value you're comfortable with, say > 80 or 85. You'd like to end up with a list of 300 or more stocks that have had good earnings growth and meet a few loose criteria. The intent is to gain an unfair advantage over the average trader, so that if there's a price setback, you can trade with a wider range because you have a reasonable expectation that you're trading a stock that is likely to recover quickly when the market recovers. If you want to do group/sector analysis, add what's needed for that (I use the S&P 500, but then again I've already disqualified myself for that type analysis) and if possible, place your first list (fundamentally sound stocks) in the correct groups.
I don't trade the AIQ ERs. I pick stocks with good fundamentals, then I trade these stocks using a money management scheme that has me add to a position or sell some of it, but never close it out. AIQ gets used to help decide where I set the next buy/sell point. I use a combination of chart patterns (standard old Magee & Edwards), AIQ bands and RSI to select the initial entry point, decide when it's time to move on to another stock and to adjust the next trade value. I usually trade 1 stock in excess of a year, a couple I've been trading for 2 years. So you can guess that I tend to look at weekly charts far more than I do daily charts.
I've wrestled with entry/exit in AIQ using indicators only and found that the recommended ERs on the Weighted Action List + confirmation (Phase + DirMov) in conjunction with market signals can work. You can expect to see a lot of busted trades that are outweighed by winners. Overall I thought I hit around the market average unless I got a big winner. When I looked back at my track record, I realized I'd left a lot of $$$ on the table. To me that meant that I was being taken out of the market for the wrong reasons and it reflected on my judgement in evaluating indicators or setting stops in the short term. The very nature of ERs make them unreliable to trade by themselves.
What parts of AIQ that you use are unimportant. They have to make sense to you. I have a friend that trades using only AIQ, he only uses AIQ bands and candles, and he trades both long and short. That's it. And he's making $$$. The only thing that he does outside AIQ is he puts some effort into what stocks he trades. There's that unfair advantage again.
The bottom line to all this is that you aren't going to solve a money-loss problem with more s/w. I think you suspected that or you wouldn't have asked if s/w was actually getting USED. AIQ can do pretty much everything that you'd ever want. It's a very complete and powerful program. They're working on their deficiencies and I think they'll eventually have the EDS tests and scans worked out. But it won't be a universal answer to technical trading... nothing is.
Bruce |
