Comments on: Boosting Strategies with MMI https://financial-hacker.com/boosting-systems-by-trade-filtering/ A new view on algorithmic trading Sat, 22 Oct 2022 11:08:47 +0000 hourly 1 By: Dan https://financial-hacker.com/boosting-systems-by-trade-filtering/#comment-62486 Fri, 27 Sep 2019 13:10:37 +0000 http://www.financial-hacker.com/?p=320#comment-62486 This post is pure gold!! Thank you so much for all the effort and for sharing. For those paying attention, there’s thousands if not tens or hundreds of thousands of dollars in this post alone. Awesome stuff… Now I’m going digging to see if i can any other experiments run over 900 different systems and found to consistently improve them!! (Would love to see one that increases the likelyhood of a mean reversion strategy working)

]]> By: Niko https://financial-hacker.com/boosting-systems-by-trade-filtering/#comment-60201 Thu, 16 May 2019 13:29:33 +0000 http://www.financial-hacker.com/?p=320#comment-60201 Ok sounds fantastic, thank you

]]> By: jcl https://financial-hacker.com/boosting-systems-by-trade-filtering/#comment-60197 Thu, 16 May 2019 07:40:07 +0000 http://www.financial-hacker.com/?p=320#comment-60197 No, but someting is obviously different. If it’s not the historical data, then asset parameters or something in the software have changed since that article from 4 years ago. I will look into that, and run the test again in the next time with current parameters. I’ll post here what I find.

]]> By: Niko https://financial-hacker.com/boosting-systems-by-trade-filtering/#comment-60184 Wed, 15 May 2019 16:06:52 +0000 http://www.financial-hacker.com/?p=320#comment-60184 I’ve tried replicating these results by placing the line “EndDate = 20150901;” in the TrendMMI.c file (in 2015 Scripts archive) and running both the TrendTest.bat file, followed by BoostrapMMI.c

Unfortunately, even with the date constraints, the “Best” system is Strategy\TrendHMA.c_XAG/USD_H4_1_1: 1.28, with a Sample median of 1.28 and a P-Value of best strategy 0.41.

Any idea as to the discrepancy between the blog article and my results?

]]> By: jcl https://financial-hacker.com/boosting-systems-by-trade-filtering/#comment-39608 Sun, 02 Sep 2018 08:15:32 +0000 http://www.financial-hacker.com/?p=320#comment-39608 Yes. Bitcoin does not behave like a currency, but trend is present.

]]> By: Mark https://financial-hacker.com/boosting-systems-by-trade-filtering/#comment-39588 Sat, 01 Sep 2018 10:12:58 +0000 http://www.financial-hacker.com/?p=320#comment-39588 Interesting results.

Have any similar experiments been conducted on Bitcoin to determine if it has similar “currency” characteristics?

Thanks,

Mark

]]> By: jcl https://financial-hacker.com/boosting-systems-by-trade-filtering/#comment-17154 Tue, 17 Oct 2017 08:09:08 +0000 http://www.financial-hacker.com/?p=320#comment-17154 Falling is simply a negative slope.

]]> By: Gene https://financial-hacker.com/boosting-systems-by-trade-filtering/#comment-17137 Mon, 16 Oct 2017 20:24:13 +0000 http://www.financial-hacker.com/?p=320#comment-17137 Hi, what is your definition of falling in the code above. Is falling testing to see if MMI is below a threshold of 75% or is falling testing that the slope of the MMI is negative i.e. sloping in a downward direction (\) or is falling testing both that both of the mentioned conditions are true. Thanks

]]> By: Darrell Washington https://financial-hacker.com/boosting-systems-by-trade-filtering/#comment-14053 Thu, 27 Jul 2017 05:30:23 +0000 http://www.financial-hacker.com/?p=320#comment-14053 Thank you very much jcl for response. Really appreciate it.

]]> By: jcl https://financial-hacker.com/boosting-systems-by-trade-filtering/#comment-14029 Wed, 26 Jul 2017 07:37:12 +0000 http://www.financial-hacker.com/?p=320#comment-14029 Swing trading is ok. I recommend time frames of 1 hour or above. Below 1 hour, it becomes difficult and takes experience to achieve profit and to properly backtest systems.

]]>