3/6 [Backtesting Evaluation Method] We explain all items of the backtesting image⑤
EA Developer Reiwa's Double-E is me.
In the first place
EA developers gauge the quality of an EA by checking with
“backtests.”
So as users
to “discern the quality of an EA,”
you need the skill to pick through backtests.
With that in mind, this series
will describe “how to find winning EAs from backtests.”
【About the period】
As for the backtest period,
it is generally expected to be over 10 years.
At least within 5 years, you haven’t experienced many market conditions.
The EA that was recently fitted may work in the short term,
but whether it can win in the long term is questionable.
Because it is aligned with the most recent market conditions,
it may capture market tops and bottoms, but
it is a double-edged sword, and
there is the fear of not knowing when it might stop working.
I consider anything with only 1 or 2 years to be out of the question.
Also,
please refer to the following article scheduled for 4/17 publication.
EA Developer Reiwa no Doubles-E here.
This time“Whether to include the Lehman Brothers period in EA development”
So,
in fact this is a topic where even EA developers disagree.
To cut to the chase,
I want to validate with many chart patterns,
so I am on the side of “include.”
However,
as the “do not include” camp,
・It is said to be once in a hundred years, so it can be excluded
・In market conditions like Lehman, EA might not run at all
・Lehman’s moves are an “anomalous value,” so exclude them
・Smartphones and FX trading environments differ now, so it’s not relevant
etc., but I do understand those points.
So, will such rough moves like Lehman’s appear again in the future?
What if I said it happened for just “one month”?
Yes, the Corona shock market.
The VIX index (fear index) at that time reached levels unseen since the Lehman era.
The dollar-yen, which was in the 112 range,
fell to the 101 range and then rose back to the 111 range,AUD/JPY and NZD/JPY reached near all-time lows.
With COVID information being chaotic,
some people didn’t realize the VIX had spiked,
and didn’t have time to stop their EA,
which probably happened to many as well during Lehman.
So personally,
I am on the side of including Lehman period.
In other words, it is also a rare opportunity to validate “movement of the century.”
If you can cover all kinds of market environments,
even without pausing the EA for some reason,
you can design to minimize large losses,
which is very important to me.
Now then!
※ Among users,
there are also people who do not include Lehman, so
some listed products include development and backtest data from 2013 onward.
With that in mind,
this wayI can clearly explain the period of EA developmentlike this,
but some developers say “excluded because they didn’t win during Lehman period.”
So, some EA developersmight say so,
If you’re curious, ask the EA developers directly in the community or via private message.
This is not fear or anything;
in the stock world, some people visit the company to decide whether to buy stock.
It’s the same thing here.
【Reiwa’s Double-E EAs Listing】
https://www.gogojungle.co.jp/users/112481/products
【Reiwa no Doubel-E Series Articles】
https://www.gogojungle.co.jp/finance/navi/series/1700
https://www.gogojungle.co.jp/finance/navi/articles/64703
https://www.gogojungle.co.jp/finance/navi/articles/64738
https://www.gogojungle.co.jp/finance/navi/articles/65619
https://www.gogojungle.co.jp/finance/navi/series/1701
https://www.gogojungle.co.jp/finance/navi/articles/64723
https://www.gogojungle.co.jp/finance/navi/series/610
【Reiwa no Double-E’s Belief】
