Anyone use APS Automatic Pattern Search to find forex patterns? I would like to hear your experiences about the best exit targets to use and the timeframes that provide the most promising patterns. The way things are going with the stock markets I think I will be concentrating on forex for a while. Alex
I tried the demo with tick data and I wasn't able to get any promising patterns. I will try it with 3 minute bars and if I get something I will let you know. Thank my super fact computer
Hi Alex, I don't use timeframes less than 15 minutes. APS appears to work best with 15-minute bars or more. The number of bars in the file needed for the search to have any meaning increases as the timeframe decreases. The company gives a formula to estimate the amount of data in the file but I use my own formula.
Thanks Bill, I know about this but my objective is to try to determine whether there is a particular timeframe in forex that provides the most potential with pattern trading. I got a fast computer and I just started an APS search run with mutliple time frames and exit parameters. It may take up to a week or so for the search. I will then forward test these patterns and repeat this for several currency pairs to see if there is a correlation between timeframe, exit levels and pattern forward performance. Timeframes: 1, 3, 5, 7, 10, 15, 20, 30, 60, 120, 240 (in minutes) Exit levels in pips: Profit target: 5 - 100 in increments of 5 Stop-loss: 5 - 100 in increments of 5 Profit target/stop-loss ratio: 1, 1.25, 1.5, 1.75, 2 Profit factor: > 1.5
I think this is a lot of work and APS can do it for sure but I only use 15 and 30 minute bars and 20 - 50 pip exits and I get many patterns every time I run the program. The difficult part is forward testing all these patterns and although the new version makes this a lot easier than it was before you still have the problem of deciding which patterns have a good chance of maintaining their profitability. I think APS is a good tool for finding patterns but my problems start after I get the output from the program. My suggestion is try spending less time looking for optimum combinations of bar timeframes and exit levels and concentrate on methods for determining probability of future performance of patterns, that is where the money is.
Bad news, the demo just expired on my face. I will have to download again to another PC. I get so pist when demo software expires like that.
I know this is going to sound a little off-the-wall but have you ever considered......paying for it :eek: I get so 'pist' when freeloaders refuse to support indepependent software developers, especially if it's a decent program which they're making money from! Mind you, $1295 for a 3 year licence is a bit much for this type of software, offer them $100 and see what they say
jimbojim has attempted in the past to make "exchanges" of software. I think that people who cannot afford to invest at least $5K in software and at least that much in hardware should not bother trading. Markets are crowded with large fish, Market is no place for small time gamblers. They are usually out of the game after a few trades. Then they try again and repeat their mistakes. When they learn they are either too old or too poor to trade, or both.