Evaluating the effectiveness of filters in the construction of an ATC - page 6
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
...Have a look at the other sheets, please, the filtering principle is the same and it's very interesting, given the conclusion from the weak filtering that in fact the filter doesn't work - well it doesn't :)
And, you are investigating an EA that trades against the trend with averaging - so, in my opinion, evaluating it by profit is not very objective.
I'll see, there will be time... And what is objectively considered a performance indicator? By the way, a couple more words about the first filter. It cuts off the tails of the distribution, with both tails: the worst results and the best ones. This is not good either. Ideally, the filter should cut off the left tail - the worst ones.
I have just decided to give myself an answer to this question - I sat down to write in Word, to make logical calculations, but I still come to the conclusion that the net profit is not a good indicator, at least for the evaluation of such TS. Although I don't know how to post my note about it - I've already got 14 pages long, but I want to examine my position critically, especially concerning methods of chart analysis...
As for cutting ends, I think you need to look at how much profit and loss decrease in percentage terms. But, for me personally, the number of deposits lost during optimization is very important - the less deposits, the more chances to survive. It's better to live poorly but live poorly than to have 1 in 100 chance to get rich...
As for cutting ends, I think you need to look at how much profit and loss decrease in percentage terms. But, for me personally, the number of deposits lost during optimization is very important - the less deposits, the more chances to survive. It's better to live poorly but better to live than to have a 1 in 100 chance to get rich.
It's a matter of money management when you lose deposits.
As for the filter - a uniform growth of TF with a uniform decline in total profits is good)) where is the compromise, it depends again on money management. And in general, you can use it to trade a portfolio of one system with different settings (including filter parameters). Simply, if the signal triggers with a better filter value, the lot is larger, and with a weaker one, the lot is smaller. Of course there is a PF threshold below which there is no point in trading the system. Similarly, you can combine the system with different settings that will result in staged entry and/or exit, which may be a diversification in some sense.
losing deposits is a matter of money management.
The seventh filter is a strong cutter. But there is a nuance. It's skewed in the right direction: it cuts the left-hand tail more.
This can be seen from the distribution. The Gumbel Max was the most suitable.
The important conclusion is that about 30% of the passes were unprofitable and about 70% were profitable.
The deposits we have lost are a matter of money management.
As for the filter - uniform growth of PF with a uniform decrease in the total profit is good)) Where is the compromise, it depends again on the money management. And in general, you can use it to trade a portfolio of one system with different settings (including filter parameters). Simply, if the signal triggers with a better filter value, the lot is larger, and with a weaker one, the lot is smaller. Of course there is a PF threshold below which there is no point in trading the system. The same way we can combine the system with different settings that will result in staged entering and/or leaving, which may be regarded as diversification in some sense.
Much depends on the TS, if it is the so-called averaging, as in the example, then the loss is equal to a failed entry, and the fewer failures, the better - for this is responsible filtering system.
If I may give an example, I do not immediately understand the phrase " uniform growth of PF with an even decline in total profits is good" growth in relation to what - the last pass? Then what is the overall decline in profits?
The second filter is the same as the first one - it cuts tails left and right...
And what kind of result do we expect - here we would like to understand the benchmark against which, theoretically, we are trying to determine the effectiveness of the filter.
The seventh filter is a strong cutter. But there is a nuance. It is skewed in the right direction: it cuts left-hand tail more.
This can be seen from the distribution. The Gumbel Max is the most suitable.
The important conclusion is that about 30% of the passes were unprofitable and about 70% were profitable.
In practice I use 3 and 4 settings of the same filter.
Is 30% and 70% conclusion relative to results of the F_7 sheet? Just look at all these sheets and you will see the following situation
The asymmetry coefficient for the Humbel distribution is constant at 2.404 - what does this give us theoretically - what is the point of this model?
A bold statement. Imho, MM is Dr Bormental (assistant), not Professor Preobrazhensky (chief medical officer).