| Pages:
1
..
7
8
9
10
11
..
47 |
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
I hope to show a new method of building systems.
Here is the advantages
Common sense shows it should work well. (GSB sometimes shows us our common assumptions are faulty)
Lower CPU power needed.
Higher proablilty it will work out of sample.
Human time to build a system is not to extensive either.
GSB needs some tweaks to improve the process, but it still can be done as is.
In summary, we test 1 indicator at a time. Then with this one indicator forced to be used, we rank the other indicators by fitness.
We look at the top 1/3 and bottom 1/3 -In the nth out of sample, and the dates after 20160630 (nth all)
This is what results clearly showed.
One indicator is clearly top, a number are very good. A number are in the top 1/3 and bottom 1/3. I wont use these.
A number are great in sample, and poor out of sample.
From here we run GSB with say the top 6 indicators. Then we choose the systems with best OOS metrics.
|
|
|
rws
Member
Posts: 114
Registered: 12-6-2017
Member Is Offline
Mood: No Mood
|
|
That's a good idea. A bit similar to automatic features selection in machinelearning. But there is a drawdown.
Sometimes input A and input B separately analyzed could show they don't work but while having A and B together they could work very good.
I still think it's a very good idea.
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
Quote: Originally posted by rws  | That's a good idea. A bit similar to automatic features selection in machinelearning. But there is a drawdown.
Sometimes input A and input B separately analyzed could show they don't work but while having A and B together they could work very good.
I still think it's a very good idea.
|
Thats a valid point, but we can start with one indicator.
Then try combinations of two indicators.
Im excited about this, often research is painful slow and boring (till you strike gold), but this is going to be quick and informative.
We can also pick the top indicator, exclude it and look for the next one.
or we can pick the top, keep it and look for the next best ones.
Love to hear how this works for people.
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
Ive spend a lot of time testing what indicators are best and worse. The process is fairly quick in human and cpu time.
Results so far are more erratic than I expected, but show definite bias to some indicators being very good, and other poor.
Still have not perfected the methodology. We can make lots more systems that are similar to get a bigger sample size, by using operand + and making
the weights on all but one indicator as zero. Any indicator multiplied by zero is zero. GSB is not smart enough to release this so we can then get
multiple systems that are very similar. Normally GSB filters out similar systems
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
I hope I have now perfected how to determine what indicators are best and what isnt. Results now are consistent each time. Doing one more day of
testing with a bigger sample size to confirm. Systems should also build faster as we can drop 1/2 the indicators.
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
Here is the summary on es.30
Closedbpv was excluded from all tests as its in this case the same as closed.
I will publish the best indicators in the private forum in a day or two.
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
For ES, using 2 indicators and operand +, this is a cut down list of what indicators work and dont.
I will publish the entire list in the private forum next week.
GSB is getting a modification to allow systems that are very similar, but not 100%. This is needed for me to get a larger sample size
using only two operators and multiply.
I will also publish the excel spread sheet too. But I'm waiting on GSB to have custom coulombs to be exported.
What Im interested in also is does the list vary when we use * instead of +
and will the list vary on other markets.
What I hope the cut down indicators will do is,
1) Build systems faster, possibly use less ram??
2) Make more robust systems
3) Give better our of sample results.
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
Now that the export of GSB indicators is customized, I will upload the excel template I use to work out the stats on each indicator.
Will be done < 24 hours.
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
Here is the template. Feedback welcome
Attachment: Login to view the details
|
|
|
Bruce
Member
Posts: 115
Registered: 22-7-2018
Location: Auckland - New Zealand
Member Is Offline
Mood: No Mood
|
|
Here's what I got for NQ30, does this look reasonable?
Not sure why I got the #DIV/0! on row 16. halves the number of effective indicators!
Will do a new build based on these findings and see what comes of it.
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by TradingRails |
Here's what I got for NQ30, does this look reasonable?
Not sure why I got the #DIV/0! on row 16. halves the number of effective indicators!
Will do a new build based on these findings and see what comes of it. |
I suspect you dont have bol enabled. Just released you need both of them enabled
Upper and lower
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
Sorry I miss-read what you wrote tradingrails. Check ct2 is enabled
|
|
|
Bruce
Member
Posts: 115
Registered: 22-7-2018
Location: Auckland - New Zealand
Member Is Offline
Mood: No Mood
|
|
As I used only the * operator for this analysis, is it ok to use both * and / operators for the system build?
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by Bruce |
As I used only the * operator for this analysis, is it ok to use both * and / operators for the system build? |
I don't think there is much upside to "/"
Also its more problematical with the possible multiplier values
two indicators has max range of -100*-100 to 100 * 100
ie -100,000 to 100,000
with / it could be the same, but results could be -1 to 1 if / is used.
Then Entry parm's should be say -0.5 to .5 step .1
It would be valid and ok to just make entry parm 0 for all tests.
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
I have had a significant improvement in the process.
My results with 4000 to 6000 tests were fairly erratic.
I did 2 identical tests, and got 15 and 16 indicators >=1
However 13 were different in each test. Thats not a good result.
Thought the highest few were the same.
After doing 3 identical tests of 40,000 to 50,000, I got two tests results identical, and the numbers very close.
The third test differed by two indicators. Difference in results was -.12 and .19 - which is not large.
This was using *, 2 indicators and duplicatesystems set to true
|
|
|
Bruce
Member
Posts: 115
Registered: 22-7-2018
Location: Auckland - New Zealand
Member Is Offline
Mood: No Mood
|
|
I've performed a lot of testing with this new process focusing on NQ and SI, so I thought I would share the following observations.
There's no doubt that this analytical process significantly improves the quality of a system builds by eliminating what appear to be ineffective
indicators in OOS testing. The quality of verification testing and VSS scores are also noticeably improved, all this leads to better WF results.
NQ which I have found to be a more challenging contract to build good systems for, upon performing Peters indicator test (I've published one of the
test results here) the resulting system builds were noticeably better, more systems passing VF and delivering better quality WF output.
SI has been a lot easier, really good systems, with smoother equity curves. What I'm also noticing is the number of indicators that appear
consistently to over or underperform across different assets as a result of this new testing process.
|
|
|
traderusa
Banned
Posts: 5
Registered: 24-5-2018
Member Is Offline
Mood: No Mood
|
|
That's interesting to see. How many systems were built to get this list? Peter found that results varied a fair bit on 10,000 sample, but 40,000 was
consistent.
Do you expect the same indicators to be good / bad on all markets, or market groups, or unique to specific markets?
|
|
|
Bruce
Member
Posts: 115
Registered: 22-7-2018
Location: Auckland - New Zealand
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by traderusa | That's interesting to see. How many systems were built to get this list? Peter found that results varied a fair bit on 10,000 sample, but 40,000 was
consistent.
Do you expect the same indicators to be good / bad on all markets, or market groups, or unique to specific markets? |
There were @10,000 on this particular run, I had an earlier test with only 5000 and that was nearly as good. That would indicate the more the better
the result is likely to be.
I don't expect the same the bad/good indicators across all markets however early indications are that some do appear to be more consistent than
others.
You can see you get a number that are +/- 5 or so, but it's those that are >+/- 10 that gets your attention.
|
|
|
traderusa
Banned
Posts: 5
Registered: 24-5-2018
Member Is Offline
Mood: No Mood
|
|
Thanks for the reply Bruce. Could you publish a silver GSB performance report. If any of it is sensitive, hide the trade list and or settings, but
state what slippage and commission is used. There is no other reference in the forum to anyone using GSB on silver, so this is interesting.
|
|
|
Bruce
Member
Posts: 115
Registered: 22-7-2018
Location: Auckland - New Zealand
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by traderusa | Thanks for the reply Bruce. Could you publish a silver GSB performance report. If any of it is sensitive, hide the trade list and or settings, but
state what slippage and commission is used. There is no other reference in the forum to anyone using GSB on silver, so this is interesting.
|
Attached is the silver report form TS without any Slippage and Comm accounted for. Apart from the fac that the average trade is substantial so C&S is
easily accommodated, I change the entry and exit orders to Limit Orders in most systems and use the TS function to change the limit order to a market
order after 15 seconds which in reality rarely gets triggered.
For indexes I use buy 'minlist( C, AvgPrice) Limit;' or for shorts, 'SellShort MaxList( C, AvgPrice ) Limit;' have been doing this for years and found
it to be really effective adding to the nett. There are a few other tricks as well! 
Attachment: Login to view the details
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
Great result Bruce. I had never tried GSB on silver.
Here is an example of what secondary filter works best. Its is fairly simple but critical to get right.
For stock market indices, it always is CLosedMinusClose (not GA)
|
|
|
traderusa
Banned
Posts: 5
Registered: 24-5-2018
Member Is Offline
Mood: No Mood
|
|
Can someone publish how much results improve if we use all indicators, the top 1/3 only, the top half and the top 2/3
I would like an objective comparison to see if this is worth doing.
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by traderusa | Can someone publish how much results improve if we use all indicators, the top 1/3 only, the top half and the top 2/3
I would like an objective comparison to see if this is worth doing. |
I can supply that tomorrow, including 25 30 35 min bar comparison.
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
Hi TraderUsa. Here are the results.
|
|
|
Bruce
Member
Posts: 115
Registered: 22-7-2018
Location: Auckland - New Zealand
Member Is Offline
Mood: No Mood
|
|
this is Gold!
|
|
|
| Pages:
1
..
7
8
9
10
11
..
47 |
![[*]](images/davis/default_icon.gif){kind=icon}
