Mechanisation of optimal parameter selection. Finding a common denominator. - page 19

[Алексей Тарабанов]

I wasn't asking for the front when I was yanking the indicator. I need to change lanes :)

[Le0n]

lasso:

Copied from the thread Where is the line between fitting and actual patterns?

I think this is the limit.

=========================================================

The criterion (or line) sought in this thread should not depend on the type of TS.

I cited this TS only as a clear example for everyone.

.......................

Ok. Let's set a problem from the opposite direction:

You need any available TS by any optimization in the tester (even most severe overoptimization) finally produce the following results:

1) Number of transactions in 1 year at least 250-300.

2) The mathematical expectation of at least two spreads.

3) Let the recovery factor be equal to four (minimum).

.............

Who can present a tester report with these results?

Immediately I see a forest of hands...

Ah, yes I completely forgot:

4) Testing range all available history from 1999 to 12.2010 (12 years)

=====================

If anyone can show something like this,

it would be appreciated.

PS And probably surprised. ))

Greetings, it's been a year and a half since I left this forum... and even more since I abandoned my flooded thread. Didn't drop back in out of pure schadenfreude... )))... Just might surprise the author of the most sobering reply to the thread... Or rather share the surprise.... sharing...

In addition to the report:

- This is not the result of optimisation/fitting... there is no place to optimise at all in this EA... The trigger is as simple as a spade... ))))

- Orders are market orders.

- Micro lot 0.01

- Expectation 2165 points

- No stops

- No trailers

- Without indicators

____________________________________________________________________________________________________________________________

[Sceptic Philozoff]

1. The number of trades is roughly equal to the number of 4-hour bars. With this quality of modelling (by opening prices), it is as if you have opened a trade on about every bar. But this is only an appearance, see point. 3.

2. the relative drawdown is very high, almost 66%.

3. a huge length of the maximum profitable series (488 trades) - an indication that the trading system itself is not Bernoullian, that is, most likely deals were opened in such big packs. It seems to me that this is quite risky. Perhaps this is the reason for such a large drawdown.

4. The significant imbalance between the number of profitable long and profitable short trades is surprising.

[Le0n]

Mathemat:

1. The number of trades is roughly equal to the number of 4-hour bars. With this quality of modelling (by opening prices) it looks as if you opened a trade at about every bar. But this is only an appearance, see point. 3.

2. the relative drawdown is very high, almost 66%.

3) huge length of the maximum profitable series (488 trades) - an indication that the trading system itself is not Bernoulli's, ie most likely deals were opened in such big packs. It seems to me that this is quite risky. Perhaps this is the reason for such a large drawdown.

4. The significant imbalance between the number of profitable long and profitable short trades is surprising.

- The number of deals is categorically equal to the number of bars in the testing period, the extra thousand in the history is 1000, which the terminal draws in the window before the start of the test.

- The relative drawdown only in % is terrible, and it happened in those unforgettable pre-crisis times, when the pair was hanging out insub-zero. Is it relevant 10 years later?

- The tester understands a sequence of deals closing with non-negative profit as a profitable series, in this case, the deals were opened at different times and in different directions. This series lasted for 16 weeks.

- There is an imbalance, but it is natural - from the start to the finish, the pair grew by 0.30. There are more bays.

Flattered by your attention, thank you.