Comments on: Better Strategies 5: A Short-Term Machine Learning System

By: jcl

jcl — Tue, 06 Aug 2024 11:48:39 +0000

In reply to Tom. The order does matter when you group the signals, as in the example. The limit is really 20, even when the array has more signals. You can believe me, I have programmed this function ;).

By: Tom

Tom — Mon, 05 Aug 2024 17:56:28 +0000

Thanks for your reply. Based on Zorro Help page for the adviseLong(): “adviseLong (int Method, var Objective, var* Signals, int NumSignals): var”, the var array ‘Signals’ can has more than 20 signals.

Does the ‘ordering’ of the signals within the array ‘Signals’ matter? as far as creating difference pattern groups for the machine learning? Thanks.

By: jcl

jcl — Wed, 31 Jul 2024 15:15:06 +0000

The signal limit for patterns is 20. Patterns with 5 candles are theoretically possible, since they have only 15 signals. But I don’t think that you will find any useful patterns of more than 3 candles.

By: Tom

Tom — Mon, 29 Jul 2024 23:43:03 +0000

Referencing program ‘Alice4a’ in chapter 6 of your book, if I want to use a pattern with 5 candles, where the total number of signals is higher than 20 for the adviseLong(), and I would need to store the signals in an array, does the ordering of the signals within the array matter? Should the signals place in an order such that they will be divided into pattern groups that I desired? Thanks.

By: leonardo

leonardo — Sun, 11 Sep 2022 11:02:11 +0000

This series of articles “Build Better Strategies” that was made in 2016 needs to be updated?
perhaps there are a new series more updated or as a base is still valid.
Thanks

By: jcl

jcl — Wed, 22 Dec 2021 14:08:17 +0000

Depends on asset and broker, but it should be well above 55.

By: Max

Max — Wed, 22 Dec 2021 14:00:49 +0000

How much accuracy is enough to cover spread and trading costs?

By: Zorro Trader

Zorro Trader — Fri, 17 Sep 2021 09:24:48 +0000

Thanks, Jcl, for a great post and a great blog! Unfortunately, it is obvious that the average profit per trade of this particular strategy is too low to make any money in real life. Also, it is interesting to notice that the accuracy of the DeepNet predictions is higher than the number of winning trades in Zorro. I posted something about that here: https://zorro-trader.co/zorro-trader-how-accurate-is-deepnet/
I guess sometimes the marker gremlins are eating our lunch…

By: jcl

jcl — Sat, 16 May 2020 15:15:13 +0000

I don’t know Trading System Lab and Genetic System Builder, but there are hundreds of similar tools around. So I can’t honestly say if they have any value or if the appearance in that magazine has any meaning.

By: Nacho Lorente

Nacho Lorente — Sat, 16 May 2020 14:48:10 +0000

Thanks for this series of articles.
But in relation to his comment “I got several emails asking about the“ trading system generators ”or similar price action tools that are praised on some websites. There is no hard evidence that such tools ever produced any profit (except for their vendors) – but does this mean that they are all garbage? We’ll see. ”
What opinion do you have regarding the software: Trading System Lab and Genetic System Builder.
Systems created in both softwares reached the top ten of Futures Truth magazine in 2018