3/25【Backtest Evaluation Method】We explain all items of the backtest image24
EA Developer Reiwa's Double-I-E
First of all
EA developers assess the quality of an EA through
“backtests.”
Therefore, as a user
to discern the merits of an EA,
you need the skill to evaluate backtests.
With that in mind, this series will
explain how to find profitable EAs from backtests.
Now,
we have finished explanations for all items in backtests, but
there are other things to read and understand, so I will explain them.
【Whether it is based on the opening price movement?】
Even if backtests look perfectly rising on a chart,
one factor that reduces reproducibility in live operation is
the question of whether it uses opening price movement.
※Considerations on opening-price-movement EAs
will be excerpted from past newsletters and planned articles.
※We also discuss whether they are often closed out by TP or SL.
※ From a post planned for 3/21 in the EA Beginner Course
EA Developer Reiwa's Double-I-E here.
Today,
“The sales page backtest must be “all ticks””
is the topic.
-----------------------------
About types of backtests
-----------------------------
First, I will explain the types of backtests.
There are three types.
“All ticks,” “Control points,” “Opening price only.”
These three exist.
All ticks ➜ backtest using four price values of a 1-minute candlestick
Control point ➜ backtest using the next-lower timeframe than the EA’s timeframe as well
Opening price only ➜backtest using four price values of tick data
In other words,
in the EA development phase,
for an opening-price movement EA, we develop with “opening price only.”
(Opening price movement: EA that operates only on the opening price of candles)
And
finally we verify with a backtest using all ticks.
Since we backtest with four price values of the 1-minute chart,
for example on a chart with 5-minute candles or higher,
how the high, low, and closing price are formed from the opening price is more accurate,
so
“All ticks” is the most reliable.
By the way,
on a 1-minute EA, MT4 can artificially determine the order of high and low
which reduces reliability and lowers the modeling quality to 25%.
Because of that, even on 5-minute candles or higher, it will not reach 100%,
and will be about 90%.
Additionally, for tick-data-suit (TDS) where spreads fluctuate,
reliability would be even higher, reaching 99.90%.
Now,
When submitting an EA for listing,
we also upload backtests,
usually submitting all-ticks.
Sometimes you will see backtests published as “opening price only.”
Some people think, “opening-price-movement EAs don’t differ much, so it’s fine.”
There are times when people are lazy.
(Opening-price-only backtests finish more quickly)
However,
even for opening-price movement EAs,
take profit (TP) and stop loss (SL) are settled by tick data, not opening price,
so the results change, and for EAs with many TP/SL,
you should really use all ticks.
Well then!
Quoted from past newsletters
About the “Trading History” If there were a graph showing endless trading history underneath the chart, I’d like you to look at it. First, what I care about is「Whether it is an opening-price-movement EA」. If it is a 5-minute chart, is it entering on a 5-minute basis, if it is a 15-minute chart, is it entering on a 15-minute basis? This is something I always check.If it isn’t opening-price movement, the reproducibility with backtests will be significantly worse.Backtest tick data use four price values: open, high, low, close. And in any EA’s backtest, even if you test with all ticks, on a 1-minute chart you only have four values, so whether high or low came first and then close, or the other way around, is something MT4 may simulate, which can cause non-100% modeling quality. (This is why modeling quality is not 100%). Therefore, EAs that move on ticks rather than opening price are more likely to diverge between backtest and forward testing.【About “not opening-price movement”】To explain to beginners: for a 15-minute EA, entering at the opening of the 15-minute candle, that is, at 0, 15, 30, 45 minutes past the hour, is opening-price movement. If it enters at other times (before the candle closes), it is not opening-price movement.
Next,So,
as a method to “find profitable EAs from backtests,”
I have explained it comprehensively.
From here on,
I will rewrite and post similar content found in past newsletters,
or add new posts if I come up with ideas or notice something.
That’s all for now.
━━━━━━━━━━━━━━━
【Reiwa’s Double-I-E EA Listing】
https://www.gogojungle.co.jp/users/112481/products
━━━━━━━━━━━━━━━
【Reiwa’s Double-I-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/1714
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’s Double-I-E Beliefs】