Tahoe, I have been using MS realtime all year and have developed several trading systems and continue to do so. My work is so far limited to the one minute NQ. I track the one minute, 5 min, 30 min, and daily charts of NQ and prefer to trade in the direction of the larger time frame. Unfortunately, MS tester/experts can only work with one time frame. I would like to include the larger time frame in my testing, but not possible. So I use the larger time frames visually with corresponding indicators which allows me discretion over the one minute signals. I use esignal with MS and it has worked well with dsl internet feed. I do wish there was a larger amount of one minute data available from esignal. According to Dave N., Quote.com has like two years, so it may be the way to go with MS. I do not think you can go wrong with MS for getting your feet wet in trying out technical indicators for use in a trading system. As I mentioned before, I started with Metastock when it was developed by Computer Asset Management back in 1985. Before that it was the Daily Graphs weekly chart books. Lots of chart reading! MS has plenty of indicators, and many not already in the program can be entered from the copy/paste method. Check issues at www.traders.com which is the web site for Stocks & Commodities mag....lots of info. and ideas.
RTS: Have you ever investigated the possibility of generating calculated data that would mimic the longer term time frame and overlay the shorter term charts? This is one deficiency I've noticed with MetaStock - that is, lack of easy support for multiple timeframe-based trading system development. Is it simply possible to modify the variables of the indicators using such that they would, in effect, represent the longer term time frame fairly well? In the triple screen trading system, the crossing of the 13 week simple moving average is the first momentum screen. (there are other ways to implement it, but this is one option). Since I have daily data, I instead used the 65 day moving average. When compared to the actual 13 week moving average on a chart with the weekly periodicity, I found that the crossovers are almost identical. Seems like that would work potentially for you too. Most people out here seem to favor Tradestation, but I want to stay with my Interactive Broker's account and continue to trade e-mini Nasdaq's. How successful have your trading systems been on the NQ's while daytrading? Just curious what I can expect. I would like to target 10 points per day. I wonder if that's a reasonable goal or not. Thanks for the response, TAHOE
RTS: Forgot one last thing. Are you using the latest MetaStock Professional 7.2? It is supposed to have 26 performance systems, with several targeting day traders. I just wondered if you had updated to that version and, if so, what you thought of their performance systems? One other nice thing they added is the option to run Performance Systems Explorations on up to 10 of their systems. This allows you to backtest multiple securities with multiple trading systems, to effectively tell you the robustness of the systems. Since I'm focussing on the e-mini NQ, I won't be using that feature, but I'm glad they added it. TAHOE
Tahoe, I have not investigated the multiple time frame subject to any extent other than the systems cannot address but one time frame at a time. I have good success with just the one minute system method, so for now I will stay with just that. My systems are a complexity of multiple systems that have grown over time. I would say, as has been mentioned in posts before, that having and testing your systems against a large amount of data is very, very important. I had a one minute system in the past optimized on two weeks of data that was 100% winners! Well, try it on any other data and it blew up. With 60,000 one minute data points that is no longer the problem. I treat the one minute charts just as I would a daily bar chart and have a variety of system to address the charateristics of market movements, ie trending, chopping, breakout, exhaustion, etc. So start addressing each area individually, and get the best results on a large amount of data, and once successful, move on to the next type of market movement. Soon you will have multiple systems addressing the action in the market. Mine are far too large to code all of the systems in one test window, so I use the formula section to create a formula for each system or system segment, and then enter the formulas in the test windows. I am currently using MS 7.02 since when I tried 7.2 I found an error in the Linear Regression indicators which I reported to MS in July. I was to have been fixed in the maintenance fix issued last month, but I found addtional problems with the same indicators. I use these indicators, so I could not use the release due to this limitation. I did try the PS in 7.2 on my one minute data, but did not find any one PS that gave the results I was able to get with my systems. I think they may be useful for ideas on improving an existing system by possibly using some of the code. Good luck with MS, I think with work/time you will be able to achieve success! And 10 pts/day is very achieveable!!