Welcome to NexusFi: the best trading community on the planet, with over 150,000 members Sign Up Now for Free
Genuine reviews from real traders, not fake reviews from stealth vendors
Quality education from leading professional traders
We are a friendly, helpful, and positive community
We do not tolerate rude behavior, trolling, or vendors advertising in posts
We are here to help, just let us know what you need
You'll need to register in order to view the content of the threads and start contributing to our community. It's free for basic access, or support us by becoming an Elite Member -- see if you qualify for a discount below.
-- Big Mike, Site Administrator
(If you already have an account, login at the top of the page)
No, I haven't been able to create a good working automated strategy yet. Things slowly got pushed onto the back burner for awhile. But I've been looking every now and then. I've stopped working on strategies that make hundreds of trades per day. I'm now looking at ES, where there is huge volume and where the little guy can get in and out without real competition (as opposed to CL).
However, the best I can come up with (so far) is + 4200 Net on 41 R.T.'s in the last 2 months with a maximum drawdown of only -680, based on TradeStation's data. I can't get more than those 2 months history with Range bars. Either TradeStation won't load it or it takes nearly the entire day to get it on the screen. And Backtesting is a nightmare due to the above problem.
The strategy works with 8-pt. Range bars and basically is just the price crossing a 6-bar moving average for a 2nd time, with a profit target and a stop-loss. But I just can't get enough trades to suit me. It's terribly frustrating to watch the ES go from 1072 to 1221 (which it has done in the last 10 days) and make only 3 R.T.'s, which resulted in only + 12 'profit'.
You need 2-3 years at minimum. Backtesting = curve fitting. Backtesting such a short time frame = ultra, ultra, ultra, ultra curve fitting. On top of that, 42 trades = worthless. 42 trades = ultra, ultra, ULTRA ULTRA curve fitting.
You can find several years of tick data in the main Elite Circle section, sticky posts at the top.
If your computer can't handle backtesting months of tick data, then I would stop using tick data and use minute bars instead. MultiCharts has a good algorithm to prevent inaccurate executions from unknown OHLC order, you are somewhat 'safe' using a 1 minute bar.
Thanks for the response, Mike. And you're entirely right in everything you wrote.
I didn't know we had all that data for our use on your site. I'll for sure check into it.
And my program is based on Range bars -- not time. My computer can handle the backtesting -- it is a quad-core with 8 mb of memory. It ran very well with MC. But it and TradeStation don't get along well (!).
If I can find something anywhere near profitable (that can be traded in automatic mode), then I'm back to MultiCharts for some real backtesting. Perhaps I'll try the 1-minute bars with TradeStation if it won't handle my 8-pt. Range bars over an extended period. (When I say '8-points', I mean 8 FULL points, or 32 ticks.)
Yes I understood. TradeStation is single threaded. MultiCharts will be 4-8 times faster on backtesting, probably more, on a modern quad core computer.
Add to that, Range bars are constructed using tick data. Minute bars have basically four properties per bar: Open, High, Low, Close. Whereas a single range bar has dozens, hundreds, possibly thousands of ticks associated with it depending on the market instrument and size of the bar. The disadvantage to minute charts (other than structure) is the ordering of the OHLC is not known. But like I said, MultiCharts has a safe guard that means that it will assume the worse, while other platforms tend to assume the best. This means that if you are long, MultiCharts will assume the low of the bar came after you got long, and before the high of the bar, therefore generating a loss.
Anyway, you should create a separate thread now that you are no longer using MultiCharts.