Econometrics: let's discuss the CU balance sheet. - page 2
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
why write meaningless posts?
The point is that on 47 deals all the robustness analysis on PF, MO, FS, MAKSDD is useless. And the topic starter is trying to do that
- We need to adjust the profit factor for the level of risk (Standard Deviation)
What is SD? Above I have sd for balance noise, is it sd = 14 pips? And how do you get the %? Percentage of what?
The point is that for 47 trades all the robustness analysis on PF, MO, FS, MAKSDD is useless. And the top starter is trying to do it.
Criticism accepted. I will try to post a different result.
What is SD? Above I have sd for balance noise, is it sd = 14 pips? And how do you get the %? Percentage of what?
sent it to the personal message.
read the "terms and conditions of participation" and note the "top list"
The point is that for 47 trades all the robustness analysis on PF, MO, FS, MAKSDD is useless. And the top starter is trying to do this
"There are estimation methods that work for such a small number of trades."(C)?????????????????????????????7
I think the robustness of the system is largely determined by the behaviour of the balance line error. here is the result of the stationarity test for the error.
Null hypothesis: The random component is not stationary
Exogenous: Constant
Bandwidth: 159 (Newey-West automatic) using Bartlett kernel
................................... Adj. t-Stat ...........Prob.
* Phillips-Perron test statistic -30.98050 .........0.0000
This is the point - any tests can pass because there are few trades. For example, let's take a bull market and all strategies with a predominance of longs or trends will be highly profitable. Even occasional "longs only" deals. The same is for flat areas, and for almost any fairly short period there are strategies that have earned not because they are robust, but simply a piece of the chart is good for them. There is a standard method - increase of testing period and number of deals. But it's a bit crude - extensive method)).
"There are estimation methods that work for such a small number of transactions."(C)?????????????????????????????7
there are, but I am not hired to describe them publicly
The problem is this - any test may pass because there are not enough trades. For example, let's take a bull market and all strategies with predominance of longs or trends will be highly profitable. Even occasional "longs only" deals. The same is for flat areas, and for almost any fairly short period there are strategies that have earned not because they are robust, but simply a piece of the chart is good for them. There is a standard method - increase of testing period and number of deals. But it is too crude - extensive method))
The "buy and hold" strategy and carry-trading are nervously smoking on the sidelines..........
Especially in kerytrading, it seems that you have to wait 100 years to get enough trades.
I've emailed it to you.
read the "terms and conditions of participation" and note the "top list"
I don't get it. SD was calculated before detrending or after?
buy-and-hold strategy and carry-trading - it looks like it would take about 100 years to get enough trades there..........
Especially with the kerytrading - it looks like you have to wait 100 years to make enough trades there.
If you understand the system's edge - the market logic, it allows you to reduce the number of trades on tests to put the system into action. Ideally, you could not look past and test at all. But these are not statistical methods, here we are talking about statistical