GSB Forums - Powered by XMB 1.9.11

There is a critical TS 9.5 bug on @GC. Im using exchange time 700 to 1330.
Before march 2020, we get bars before 700am
On local time the issue is fixed.
However systems made in GSB on exchange time work well. Local time work like crap in GSB. (there should be zero difference)
GSB systems will not match ts if they are made on exchange time in TS.
If systems made in gsb on local time, match with ts is perfect.

I want to compare this to MC / iqfeed
Can someone send me gc.1minute bars
time 700 to 1330 exchange time, exported in exchange time and a file in local time
dates from nov 1 2006 till today

Im interested if ts 10 has the same issue, but prefer to try mc first
Later note.
ts 10 has identical issue
This is the date at which the bug shows.

Hi Peter,
I have the same issue on other ticker symbols (TS 9.5 update 28).

Using session 0700-1500 on Exchange Time setting: but chart shows 0630-1500 between November and March. This issue returns every year in the chart. The chart data between April and October is correct.

Update
On one of my laptops I have opened TS for the last time in July this year.
Opening TS offline now shows the correct price data between 2006 and July 2020 (TS 9.5 update 28).

Is it a bug in TS software or in the data?

admin - 30-9-2020 at 05:02 PM

THanks so much for this. Things like this could kill traders profitability, and we have no idea about it. Im going to document this is gsb docs.
Its a bug in ts session time logic. Solution is to use local time. What symbols have you seen it in? I sue local time normally, but developed cl on exchange time. I need to check my data on this.

admin - 30-9-2020 at 05:04 PM

Later note. my cl data is ok. It it wasnt likely I would have to revisit all cl work again.

engtraderfx - 30-9-2020 at 10:06 PM

Peter, just emailed some IQ/MC data, it only goes back to early 2007 though, don't know if others find this but i find IQ data a bit flakey in early years around 2007/early 2008, gaps, low vol, spikes. I triy to get them to correct obvious spikes but generally exclude some of it from backtesting. Don't remember having any issues with bars outside session time though. Regards, Dave

Carl - 30-9-2020 at 10:59 PM

Hi Peter,

I did a few tests (TS 9.5 update 28 build 3344) and saw this issue on all tickers I tested: CL, RB, ES, NG, HO, TY, EC.

Session time 0700-1400
Chart shows 0600-1300 between Nov and March (this is not correct)
Chart shows 0700-1400 between April and October (this is correct)

I am going to contact Tradestation today. This issue needs to be solved quickly, because it is likely to appear again in the beginning of November and is going to affect our trading.

After you mentioned the issue here on the GSB forum, it is easy to spot on a chart, The double dashed lines show where the times are off.

NG 0700-1430 bug in time sessions.PNG - 9kB

Update Oct 1
I am pretty sure this issue is related to the summer time / winter time switch.

NickW - 1-10-2020 at 07:34 AM

Peter,

I have had some major data issues with TradeStation backadjusted contracts for the past two weeks. There is a lot of data missing, like ES missing all of March this year, etc. It is so bad that I switch to using IQFeed data. I do not know how this company is still in business. Cannot even call them anymore.

Nick

admin - 1-10-2020 at 10:48 PM

Quote: Originally posted by NickW

Peter,

I have had some major data issues with TradeStation backadjusted contracts for the past two weeks. There is a lot of data missing, like ES missing all of March this year, etc. It is so bad that I switch to using IQFeed data. I do not know how this company is still in business. Cannot even call them anymore.

Nick

I would close ts, delete your cache. Open a dummy workspace that has @es.d from the first possible date. Then once thats loaded open your live workspace.
TS is not kovid friendly, but we understand why.
My TS data is ok, apart from this one issue ive complained about

NickW - 2-10-2020 at 07:47 AM

When I had the data issue (looks like it's fixed now), i tried everything you suggested, in addition to creating a brand new server, installed TS, and refreshed data without any luck. I even pulled the data into MC from TS and same issues, on original server and on new server.

So i am eventually going to create a program to check all the data when i restart the environment on Sunday. I have lost too much money on phantom trades due to bad data in the platform at the time.

admin - 2-10-2020 at 05:30 PM

Quote: Originally posted by NickW

When I had the data issue (looks like it's fixed now), i tried everything you suggested, in addition to creating a brand new server, installed TS, and refreshed data without any luck. I even pulled the data into MC from TS and same issues, on original server and on new server.

So i am eventually going to create a program to check all the data when i restart the environment on Sunday. I have lost too much money on phantom trades due to bad data in the platform at the time.

thats a good idea,
basically make ts code to look for gap >3 days likely would do it.
Great if you publish it.

admin - 2-10-2020 at 05:40 PM

I have an idea thats a tiny but significant tweak to how we build systems.
With the two pass method.
choose indicators, every 80 days in sample, 80 days out of sample pre 1.1.2015 {exact date I may vary. New market make early, well known market like cl a little later}
Then typically 3 periods out of sample after this.
Whats wrong with this?
We train on the lucrative and volatile 2007 year (not great)
We dont get training in later years, we dont get out of sample in early years

My proposal is basically nth periods expanded, but we will just use dates
pretend we have 4 buttons to click. (exact amount user definable)
button
dates beginning to day 2020.1.1.
1) true/false. 1st lot of 80 recurring periods. Use for indicator IS
2) true/false. 2nd lot of 80 recurring periods. Use for indicator OOS
3) true/false. 3rd lot of 80 recurring periods. Validation period1
4) true/false. 4th lot of 80 recurring periods. Validation period1
additional validation period 20201.1. on-wards. (I want to keep lucrative 3.2020 out of the stats)

Comments welcome

Carl - 3-10-2020 at 03:03 AM

I have an idea thats a tiny but significant tweak to how we build systems.
With the two pass method.
choose indicators, every 80 days in sample, 80 days out of sample pre 1.1.2015 {exact date I may vary. New market make early, well known market like cl a little later}
Then typically 3 periods out of sample after this.
Whats wrong with this?
We train on the lucrative and volatile 2007 year (not great)
We dont get training in later years, we dont get out of sample in early years

My proposal is basically nth periods expanded, but we will just use dates
pretend we have 4 buttons to click. (exact amount user definable)
button
dates beginning to day 2020.1.1.
1) true/false. 1st lot of 80 recurring periods. Use for indicator IS
2) true/false. 2nd lot of 80 recurring periods. Use for indicator OOS
3) true/false. 3rd lot of 80 recurring periods. Validation period1
4) true/false. 4th lot of 80 recurring periods. Validation period1
additional validation period 20201.1. on-wards. (I want to keep lucrative 3.2020 out of the stats)

Comments welcome

Hi Peter,

So if I understand your suggestion correctly, the IS period covers the whole date range, but only 25% of the days when the other three blocks are set to "true"?
Same for OOS and VAL1 and VAL2?

If so, I think it's a very good idea!

OUrocketman - 3-10-2020 at 11:03 PM

Peter,

Trying to verify I understand what you’re suggesting. So, for example, take ES.

We could use the Indicator IS period with dates from 1/1/2000 to 9/30/2007. Build systems with nth 80 no trade, auto nth to trade. Select the non-red indicators at this point.
Run the non-red indicators from step above from 10/1/2007 to 6/30/2015. Build systems with nth 80 no trade, auto nth to trade. Select the green indicators.
Run stats across validation period 1, period 2, and OOS data. We could also examine effect of walk forward as follows:
Walk forward to beginning of validation Period #1 only
Re optimize at end of validation period #1 and update parameters into validation period #2
Don’t walk forward at all.

Am I understanding what you’re suggesting properly?

cico - 5-10-2020 at 03:14 AM

Dears, based on your opinion and experience, what is the best time to run "indicator selection process"? At the begining, when I do have all the Operators, Entry Modes, Normalization Modes, etc. enabled, or at the end of the process, when I do know exactly, which operator is best for the market, which entry mode is best and run indicator selection for this specific setup only? Thanks for your ideas.

admin - 5-10-2020 at 03:17 AM

Dears, based on your opinion and experience, what is the best time to run "indicator selection process"? At the begining, when I do have all the Operators, Entry Modes, Normalization Modes, etc. enabled, or at the end of the process, when I do know exactly, which operator is best for the market, which entry mode is best and run indicator selection for this specific setup only? Thanks for your ideas.

I do it at the beginning. The newest CL video is good to look at.
Only run * operator. Normalization mode CS is slightly better, so I only use that.
Whats important is entry mode. Either compare2 or a cross entry type is clearly the best. But it varies per market.
ES use cross, CL use compare. You ask good and important questions.

admin - 5-10-2020 at 04:23 AM

Quote: Originally posted by OUrocketman

Peter,

Trying to verify I understand what you’re suggesting. So, for example, take ES.

We could use the Indicator IS period with dates from 1/1/2000 to 9/30/2007. Build systems with nth 80 no trade, auto nth to trade. Select the non-red indicators at this point.
Run the non-red indicators from step above from 10/1/2007 to 6/30/2015. Build systems with nth 80 no trade, auto nth to trade. Select the green indicators.
Run stats across validation period 1, period 2, and OOS data. We could also examine effect of walk forward as follows:
Walk forward to beginning of validation Period #1 only
Re optimize at end of validation period #1 and update parameters into validation period #2
Don’t walk forward at all.

Am I understanding what you’re suggesting properly?

I am thinking of replacing nth with this.
Its nth can be say 80 trades In sample, 80 oos over say the first 70% of years.
this can can 80 day IS, 80 days oos1, {used for indicator selection} 80 days oos2 80 day oos3 re-occurring.
We would do indicator selection on first 80 is) and second 80 as indicator oos
then build systems on these 80iis1, 80 iis2, and use 80iis3 & 80 iis4 as validation

and then maybe use 2020 as another validation period

cico - 5-10-2020 at 04:33 AM

Peter, thanks for the answer. I saw the CL video, you are doing indicator pass with 1 operator, 1 entry mode and 1 normalization mode. I'm more thinking to try to do indicator pass with full set of operators, full set of entry modes and full set of normalization modes (to diminish potential overfitting). Or do you think it is not an ideal thinking? :-) Thanks.

admin - 5-10-2020 at 04:37 AM

I think its defiantly not a good idea. Your welcome to test the identical setup, and look at the OOS results. However to be precise, all tests have to be done 4 times as there is a lot of variation of identical tests. I expect OOS to be well down with what you propose.

admin - 9-10-2020 at 05:59 PM

Here is the update of the week. A lot of babies being born at the same time, and so I am the bottle neck.
1) GSB automation now has the ability to shuffle through session times. So now im testing gold from 00 to 730 in 30 min session time increments.
I will release this build late next week. Likely the last update before official release. Its a fair bit of work for me to document and to the video on in.
2) A new GSB will come out needed for automation. It has walk forward fix for ewfo, and might have 2 new indicators. Median and closesup&down
Why I might not is they have not turned up as used indicators in the testing I did on gold.
3) I am working with Murray Ruggiero a bit and some of his ideas may make there way into GSB.
Much time was spend looking at ZonetraderES by Murray. Whats significant is the loosing periods trades to GSBsys1ES is about -0.3. This is amazing considering its day trading the same market
and it has a trend following and counter trend system in it. Whats that got to do with GSB? Right now nothing which is whats so exciting.
ZT has no oscillators apart from ATR, and is mainly focused on pivot points.
IF countertrend / retracement architecture makes its way into GSB, that has lots of great implications. I may with Murray also sell my variation of Zonetrader2 as well. It has simplified logic, more symmetrical code
0.43 correlation to the original ZT, but most interestingly - trades a fair bit more often and makes more $. Whats stopping me is just investing a bit more time. Took at least 2 days work for me in the last 2 weeks and I would like to do more work on parameter choice.
4) Alertmon monitoring is 99% ready for release, but much more work might go into this. Cost of project blew out 10 fold to what expected, but even if was just for me, the cost savings in execution errors was worth it.
Today without monitoring, a freak set of circumstance would have made me miss out on a 2k trade. I'm only intending to release this on the virtual computer server Im offering
5) VM service. This has had some very time consuming, lengthy issues to resolve, but the backup server has been going well for a few months. Primary final server I hope to have running in 2 to 5 weeks.
Im exciting about the final product
6) My CL system thats in my CL video https://www.youtube.com/watch?v=QYg5eH2Q1_s (at 3min) is now trading at Striker securities (and their broker network). I released it instead of the newer CL system in the private forum as it has a longer track record, though its not made in the very newest methodology.
https://www.youtube.com/watch?v=QYg5eH2Q1_s

Note that ZT2 trades more often.
Note the CNTb4 entries. These are the counter trend ones, and its trading the exact opposite of what GSB would do.

admin - 12-10-2020 at 07:39 PM

There are updates to the methodology in the pipeline, but will take some time to finalize.
Automation and macros can now delver the final systems to be trades in favorites.
The count of how many systems pass and make it into favorites is also being added into the export stats file.
Happy to say this is because GSB users think of some good ideas that i Didnt

cico - 13-10-2020 at 07:33 AM

Dears, does it make sense from your experience to build ES systems with different Secondary Filters (even not the best ones)? An example from CL is showing that it could be interesting from systems correlation point of view.. Let me know your thoughts, please. Many thanks

Carl - 13-10-2020 at 10:25 AM

Hi Cico,

Welcome to the GSB forum.

Results on ES are best when secondary filter is non-normalized cloeslessprevclosed, but results can also be reasonable good (and low correlated) when secondary filter is normalized cloeslessprevclosed.

cico - 13-10-2020 at 11:25 AM

Hi Carl, thank you very much for welcoming me and for your answer. By normalized and non-normalized you do mean Secondary Filter CloseLessPrevCloseDBpv and CloseLessPrevCloseD secondary filters? Thanks for clarification and have a nice rest of the day.

admin - 13-10-2020 at 04:04 PM

Hi Carl, thank you very much for welcoming me and for your answer. By normalized and non-normalized you do mean Secondary Filter CloseLessPrevCloseDBpv and CloseLessPrevCloseD secondary filters? Thanks for clarification and have a nice rest of the day.

IF SF, is on GA, then the 46 to 106 sf can be used - all normalized
if its on Any other option, its not normalized.
Unless your a GSB guru, on ES I would only use closelessPrevCloseDBPV (not normalized)
Auto selects whats recommend for the contract.

admin - 13-10-2020 at 04:06 PM

Bruces method (which we are experimenting with) ultimately puts all the systems from the top 250 favA into FavB.
With GSB automation, we want to track how many systems made it into favB
This can now be exported in the export stats file

Bruce - 13-10-2020 at 07:51 PM

Bruces method (which we are experimenting with) ultimately puts all the systems from the top 250 favA into FavB.
With GSB automation, we want to track how many systems made it into favB
This can now be exported in the export stats file

The Key to this is to capture all OOS dates as one period and then filter to FavB. Creat Family, then walk forward that cohort and then review.

This has delivered some very good GC, US, YM, ES, NQ and SI systems which I'm all trading live.

The attached equity chart is from a wee dev fund I use to test systems before I put them into production, hence its a little ropie to start with however you can see over the past six weeks the impact I'm currently experiencing trading Live with a few of these newer systems.

Screen Shot 2020-10-13 at 8.19.57 AM.png - 53kB

admin - 13-10-2020 at 08:38 PM

Hi Bruce,
is that a sim or live ts account?
Is that account gc,us,ym,es,nq in the one account?
Interesting that it seems equity took of in recent times, and not march 2020.
Were all the systems live at feb 2020, or staggered over time as you made them?

Bruce - 13-10-2020 at 08:54 PM

Hi Bruce,
is that a sim or live ts account?
Is that account gc,us,ym,es,nq in the one account?
Interesting that it seems equity took of in recent times, and not march 2020.
Were all the systems live at feb 2020, or staggered over time as you made them?

This is a LIVE account NOT SIM trading es, nq, gc. US and YM has just been added for the live test.
prior equity is irrelevant as it's not the same systems or same build process and I've been trading other systems from other build platforms. These are all GSB builds as described.

admin - 14-10-2020 at 05:58 PM

Hi Bruce
I think your come up with some innovative tweaks.
Some of what you have done could be applied to the existing methodology, but regardless there is likely a month or much more work for me.
Super high level, we can go straight from systems build by automation to live trading.
Fav A contains 300 systems ranked only by out of sample performance. (20170630)
Bruces macro put the top systems from favA into Favb. I then put the dates back to start dates, and did wf
Note we had 13 members in family1, 11 in family2
SO we want systems to have big families = robust regardless of parameters.
I want wf results (brown curve) to be good. ie not tank in the last few years.
If your new to GSB, stick with whats already documented. Gold and methodology tweaks will take some time.
Big picture is system building once we have the correct setup is going to be very low in human time.
Getting the correct setup takes lots of time - but once we have found it, the info is shared to GSB users.
On that topic, if you have free cloud power, please send me your share keys. This is very cpu intensive
Those that contribute cpu power, get in-depth results of test.(Like we did with CL) But this will take me some time.

Carl - 19-10-2020 at 06:46 AM

Update October 19 2020
This bug seems to be solved today!

admin - 20-10-2020 at 08:20 PM

Thanks for update Carl.
After a few weeks work, I feel I have started to crack the Gold market.
Basically Bruce has improved on my ideas, but still follows the methodology with tiny but significant tweaks. Each tweak load into GSB automation and compare the results.
Bruces GSB results were starting to look reasonable, and hes live trading his GC systems I think with good results.
However even after copying his entire GSB folder, I still couldnt match his results. Right now still dont know why.
So possibly ive got something wrong. But now I think I have surpassed Bruce, and there is just so much more to tweak.
The most critical things to get right im still not decided in.
1) session time
2) do we skip the first hour(s) of the day and not allow entries
3) what secondary filter to use
4) what entry type to use?

admin - 23-10-2020 at 10:59 PM

IM doing a very big run of automation on gold. Numerous users have donated workers, but im happy to get more.
There is about 48 hours of work qued for the 200 or so workers I have.
I share results with those who contribute. Email me a share key if your interested. peterzwag@gmail.com
I will start on the automation / gold video next week.
Im also offline all sunday australia time. (day of rest so no work or looking at emails etc)

Bruce - 24-10-2020 at 10:49 PM

I haven't looked at YM for some time, utilising the #MOVE session list within AU and a few build tweaks along the way this is looking really promising.

Screen Shot 2020-10-25 at 5.45.00 PM.png - 166kB

admin - 25-10-2020 at 04:32 PM

Quote: Originally posted by Bruce

I haven't looked at YM for some time, utilising the #MOVE session list within AU and a few build tweaks along the way this is looking really promising.

Hi Bruce
can you share more details. you are having good success on a lot of markets that were harder for GSB.
Unrelated. I have done a lot of tests on gold. Today I need to summarize the results and get in my head what works. Thanks to all those users who contributed thier cloud power. Likely I will do another set of tests on gold later today

meldinman - 26-10-2020 at 07:16 AM

Continuing my discussion with peter in another thread on my research into new markets as well as using other data sources like PCVE on es. Something I have noticed is the amount of methodology and testing that has gone into finding robust entry signals as opposed to exits. In my time prior to GSB I learned that optimizing exits and trade management can really take profits far however with GSB it seems the time is spent trying to optimize good entry parameters and then most systems hold till end of day. is there a better methodology that we can build similar to entry techniques to be applied to exits and trade management? I know that holding till eod for day systems can often be the most robust technique but I'm interested to hear what other techniques people use to test through different post entry methods. Better to test management techniques post build or prior to market validation? aggressive management to limit risk or building portfolios of uncorrelated systems to diversify against risk? etc.

admin - 26-10-2020 at 03:37 PM

Continuing my discussion with peter in another thread on my research into new markets as well as using other data sources like PCVE on es. Something I have noticed is the amount of methodology and testing that has gone into finding robust entry signals as opposed to exits. In my time prior to GSB I learned that optimizing exits and trade management can really take profits far however with GSB it seems the time is spent trying to optimize good entry parameters and then most systems hold till end of day. is there a better methodology that we can build similar to entry techniques to be applied to exits and trade management? I know that holding till eod for day systems can often be the most robust technique but I'm interested to hear what other techniques people use to test through different post entry methods. Better to test management techniques post build or prior to market validation? aggressive management to limit risk or building portfolios of uncorrelated systems to diversify against risk? etc.

Bruce has done a bit of work on this with parabolic exits. I do very little with exits as my focus is on entries. Ive not had much improvements with exits. I do sometimes use extremely large profit targets. is $6000 on ES 10,000 euro on dax.
You can also play around with exits in TS/MC/NT after a system is built by GSB

Bruce - 26-10-2020 at 06:31 PM

Hi Peter,

Just wondering as we are just using 30min bars for the AU Indicator testing that has now grown to all the #MOVE functions, should we just use 30-minute bars exported from TS or still build the bars from 1-minute data, what do you think?

admin - 26-10-2020 at 06:46 PM

Quote: Originally posted by Bruce

Hi Peter,

Just wondering as we are just using 30min bars for the AU Indicator testing that has now grown to all the #MOVE functions, should we just use 30-minute bars exported from TS or still build the bars from 1-minute data, what do you think?

Im using 1 min myself, bit both will work.
as long as the bars exported from TS contain all data in the sessions used in AU

meldinman - 26-10-2020 at 07:43 PM

I've thought to use other exits post build like chandalier or key reversal etc. Didnt think it would be as robust and testing is alway much better in gsb obviously. do the eixt modes in GSB use an indicator input to calculate an exit signal?

admin - 26-10-2020 at 07:50 PM

I've thought to use other exits post build like chandalier or key reversal etc. Didnt think it would be as robust and testing is alway much better in gsb obviously. do the exit modes in GSB use an indicator input to calculate an exit signal?

the 8 exit modes use reversal and or signal disappearing of either the secondary or primary filters.
There is so much to be explored using GSB that makes a massive difference.
Bruce is making good headway in areas others have failed.
Improving exits for day trading, is never going to be a game change in my opinion.
So I personally would focus on the majors, not the minors.

meldinman - 26-10-2020 at 08:33 PM

fair enough. There's definitely plenty to be explored, many areas of which I'm still figuring out. Would be interested in hearing what you believe are more helpful areas to be improved upon. However I am more interested in optimizing for exits when it come to building overnight systems. I have gotten decent fitness in some markets however most of these systems have large leeway in giving back profits which is quite different from systems that I had personally developed in the past. Thats where I would like to explore. Doing some work with moc on losing position and D-D[1] with better results. further testing needed...

admin - 26-10-2020 at 09:54 PM

fair enough. There's definitely plenty to be explored, many areas of which I'm still figuring out. Would be interested in hearing what you believe are more helpful areas to be improved upon. However I am more interested in optimizing for exits when it come to building overnight systems. I have gotten decent fitness in some markets however most of these systems have large leeway in giving back profits which is quite different from systems that I had personally developed in the past. Thats where I would like to explore. Doing some work with moc on losing position and D-D[1] with better results. further testing needed...

I have done little with over night systems on GSB, but in this context exits are more important that day trading. You could try to build GSB systems, and add exits from your non GSB systems as a test.
Depends greatly on the market, but session start time, time of day to enter, secondary filter to use, entry type are the basic things to look at. On gold I have figured some of these out. Every market has its own nuances.

admin - 1-11-2020 at 10:20 PM

Here is a tip.
GSB on defaults is not very likely to make a retracement or count trend system.
If indicator 1 is set -2.5 to 2.5 step 1, negative weights will be tried
Note is vip we do not allow zero. The reason is x^0 = 1. this means gsb systems would be full of indicators that do nothing.
w1=1
w2=-1
so if result=rsi(30)*W1+rsi(5)*w2, this will give a buy signal on a pull back.
I know this is over simplified and we use * operand, not +

For the concept of enter long only on @es at 1700 (exchange time) till 2400, this setup should be used. (moc 1500)
potentially all indicators or 1 of 3 allowed to be negative.
I have not tried this myself on @es, but want to share the concept.
In the big picture, you need to try and make a benchmark, then tweak these weights
shown in cyan is parameters that can give negative. It orange is normal defaults

Carl - 2-11-2020 at 02:19 AM

Could this work?

Build strats, no stoploss, no costs, no slippage, let all strats through (so no filtering on results).
Then choose the worst results and terrible equity lines, and flip the trade direction?

admin - 2-11-2020 at 02:50 AM

Could this work?

Build strats, no stoploss, no costs, no slippage, let all strats through (so no filtering on results).
Then choose the worst results and terrible equity lines, and flip the trade direction?

you would want a pearsons of say -0.95 or so.
I dont see the point of the exercise but many great discoveries in life are made by accident, so you could always try it.
you are going to get millions of systems in gsb interface
You would need filters that say pearsons <0.95 etc.
so that cant be done in gsb now.

admin - 4-11-2020 at 10:45 PM

IM still working on foundation to Gold trading longer than expect.
Reason is i'm still finding new things that are useful.
Today I saw a very unusual secondary filter and a unusual session time worked well in the Favorite B stats. (via GSB_Automation updates)

I have made 60% of the content for the next video, but done no video recording at all.

Ive made a tweak to correlation. Starting all reports at 2007.1.1 This gave mild increase in correlation as some systems started years earlier then others.
If system1 starts years earlier than system2, correlation is lower artificially. Best correlation practice is to have systems start at same date.

admin - 5-11-2020 at 03:54 PM

I have great gold systems now, but not finished the foundation. Testing 50 secondary filters was done over night.
Once the foundation is done on gold, finding systems is really easy and fast
It paid of. Here is today's fills.
results includes $40 r/t slippage and commission

Attachment: Login to view the details

meldinman - 5-11-2020 at 06:53 PM

nice development peter. Interesting how low the negetive months corrolations are. I'm curios as to what sessions you settled on. It seems like your first system was built on a different session time? Does that help lower correlations?

admin - 5-11-2020 at 07:13 PM

nice development peter. Interesting how low the negetive months corrolations are. I'm curios as to what sessions you settled on. It seems like your first system was built on a different session time? Does that help lower correlations?

Correct on all points.
2am to 3am to 1230 is the best range.
Ive found one sf that loves 630 am start time.
right now im testing 40+ secondary filters on 630 am. Thats going to take about 12 hours doing 2 tests that are 1 pass indicators,then build

admin - 5-11-2020 at 07:14 PM

whats also going to help correlation greatly is completely different SF.

meldinman - 5-11-2020 at 07:27 PM

very interesting, how did you arrive at those sessions. I've tried to use sessions based on the most active times of day, however I was looking to explore using other sessions like first morning only or afternoon sessions to help break up correlations. Does automation help you test those things faster?

admin - 5-11-2020 at 10:52 PM

very interesting, how did you arrive at those sessions. I've tried to use sessions based on the most active times of day, however I was looking to explore using other sessions like first morning only or afternoon sessions to help break up correlations. Does automation help you test those things faster?

Thats one of the major points of automation.
This is 4 identical tests on each time frame used.
Takes me a few minutes to do such tests, but a lot of cpu time.
Im guessing 8 hours with 300 workers.

meldinman - 6-11-2020 at 07:18 AM

thanks for the graph here. so did you do similar testing of sessions when you changed the SF? or was this after you arrived at DPBV being your SF of choice? Also in automation is it going through the whole wf and verification process on each benchmark pass?

bizgozcd - 6-11-2020 at 07:35 AM

Peter, is this Exchange time or local time? I have a few decent gold systems that are 830-1430 Exchange time that were built in Sept before I was using AU session testing. I'll need to revisit them eventually after I finish EC.

admin - 6-11-2020 at 08:10 PM

Quote: Originally posted by bizgozcd

im using central usa time
so in exchange time its 3am-4am to 130 pm
Regardless there are mutiple valid time zones, and im still researching this.
Ive made some very interesting finds on gold this last week.
Im just about finished my testing of the foundation of what works, and now compling the content of power point slides. 80% done the material for the video, but not started recording the video

admin - 9-11-2020 at 12:09 AM

Happy to say I have done 80% of the first of 3 video on Gold and GSB Automation. The video however has not been edited at all yet.
It will be about 30 minutes long

admin - 10-11-2020 at 04:44 PM

video fairly close to done, but im stuck. ts data feed stoped at 240 pm central usa time, and there is no phone support or chat support from ts.
cant phone them... que ie close.
Anyone else have this issue?
I know one other user who does, and one who does not
if you have the issue, please post here
https://community.tradestation.com/Discussions/Topic.aspx?To...
and let me know

admin - 10-11-2020 at 05:08 PM

video fairly close to done, but im stuck. ts data feed stoped at 240 pm central usa time, and there is no phone support or chat support from ts.
cant phone them... que ie close.
Anyone else have this issue?
I know one other user who does, and one who does not
if you have the issue, please post here
https://community.tradestation.com/Discussions/Topic.aspx?To...
and let me know

after a few hours, issue resolved.....

meldinman - 10-11-2020 at 05:24 PM

funny, I had a similar issue then as well. seems to have been resolved...

admin - 11-11-2020 at 12:59 AM

funny, I had a similar issue then as well. seems to have been resolved...

please comment and or complain on ts forum.
Should it happen again, we are in big trouble as no fix on our side.

admin - 11-11-2020 at 01:00 AM

next video on GSB automation and gold is out. Please like the video, and comment on youtube also would be good.
https://youtu.be/VtxKY5saIR8
31 min long

admin - 12-11-2020 at 06:09 PM

I have sent out the conclusion of gold testing so far to the contrubutors of the project.
A big thank you to all of you.
If you fell of my email list, please email me.
Im now doing tests of 108 secondary filters. Glorious to have this all automated... all i need to do is wait... and few minutes work in excel.

admin - 19-11-2020 at 08:21 PM

Here is something im testing.
I choose 91 out of 110 or so indicators.
This was done by choosing all indicators, and take the most used ones.
Note: using all indicators is very slow, as there were at least 2 that did not work at all, and a few that don't match.
These are getting fixed or removed.
so if you look at this list of how often the indicators are used in the top 300, the least used indicators (lowest to be used, 27 of 91) were only used 10% of the time.
(Total indicators used was 884)
so if I don't use the bottom 27 indicators, how does this affect the stats?
Well stay tuned,and I will publish the results
This is done via gsb automation, and the newest visual basic spread sheet that comes with it.
Earlybird GSB automation discount expires midnight November 20 central USA time $600, then $700

admin - 20-11-2020 at 07:37 PM

Likely im going to make a new optional but highly recommend step in the methodology on indicator selection.
The big question is, what should the initial indicator selection be?
46, 110 or other combinations.?
in the above example we got bcd 6.65 with 123.25 systems in fav B
I now have this up to bcd 9.05 and 171.5 in fav B
It will be a few days of testing, but the initial results look very clear, and the theory behind it makes common sense when you think about it
do not user fisher, fisherinv (they give 0 and invfinate in ts)//using these I think slows gsb down greatly
percent (not match)
parabolic not match

admin - 23-11-2020 at 03:55 AM

my new method works significantly better on GC 200,230,300 to 1230 with closedbpv secondary filter
and gc 630 TO 1230 WITH a number of other secondary filters.
Hope to publish more in a few days.
basically if you build with 117 indicators with a fair bit of variation.
ie i did 4 tests with 3 session times, closedbpv sf and 4 tests with 6 sf
you then use the indicators used in favourites A to build your systems.
the list is different for the 2amish session time vs the 630 am session time
630 am had much bigger list of indicators, likely as there is much more variation due to varied sf used at 630

meldinman - 24-11-2020 at 09:28 AM

GSB automation has vastly improved my workflow. Working On a GC portfolio with the help of peters macros. seems like a good time to be trading Gold with volatility in gold starting to rise.
So far up to two systems with low negative correlation. shuffling through sessions and SF indicators makes all the difference.

admin - 24-11-2020 at 03:49 PM

GSB automation has vastly improved my workflow. Working On a GC portfolio with the help of peters macros. seems like a good time to be trading Gold with volatility in gold starting to rise.
So far up to two systems with low negative correlation. shuffling through sessions and SF indicators makes all the difference.

great your work seems to confirm mine. Its unusually easy to make low correlation gold systems. Going to get better with two session times and multiple secondary filters for the 630 (central usa time) ones

Carl - 6-12-2020 at 05:21 AM

See TS forum.
https://community.tradestation.com/Discussions/Topic.aspx?To...

Maybe the local time/exchange time chart issue from two months ago is back again. For more information, see earlier posts in this thread.

I checked and my price data looks fine (@ES and @CL), so no issue here.
But check your own charts to be sure.

admin - 6-12-2020 at 06:35 PM

See TS forum.
https://community.tradestation.com/Discussions/Topic.aspx?To...

Maybe the local time/exchange time chart issue from two months ago is back again. For more information, see earlier posts in this thread.

I checked and my price data looks fine (@ES and @CL), so no issue here.
But check your own charts to be sure.

I checked this issue and still have the identical TS issue to what KevinKdog posted

admin - 7-12-2020 at 06:48 AM

not really the correct place to post this, but interactive brokers is down. status page says its ok, but ib rep said thats part of the problem, they cant access it
They have no idea when it will be resolved

meldinman - 7-12-2020 at 07:53 AM

having issues still, hope it is resolved prior to the open

admin - 7-12-2020 at 11:03 AM

having issues still, hope it is resolved prior to the open

I KNOW OF 4 ACCOUNTS THAT are working, but one still is not.
1 of my 2 accounts work

admin - 7-12-2020 at 01:20 PM

both my accounts work now. Later today Im going to have a outage thread to post these issues. Please subscribe if your interested

admin - 8-12-2020 at 11:43 PM

60.44 released.
More gsb functions in eld included
counter trendhlv2 & counter trendhl2v2 The old versions had a bug and likely would not work
Countertrendmedian added

new-ct.png - 14kB

admin - 15-12-2020 at 03:51 PM

im back from leave
Heres some photos
workers for gold project appreciated for those who are contributing and shut them down of recent.

Im also adding into next build GSB all the counter trend- not normalized.
No idea if it will work or not

admin - 16-12-2020 at 05:38 PM

Ive been looking recently at the settings users are using on GSB.
I used to do this regularly, but have not done so for some time.
Here is what I saw
40% of users were doing great
10% should have made some small tweaks.
50% look like they have little idea of what they are doing.
I think it was in the First gold video, I spoke on what happens when I have a diesel car, the wife a petrol car, and we put diesel and petrol in both cars.
This is the equivalent to the user who just goes clicking settings, thinking they know what they are doing - but do not know what they are doing.
The default behavior should be, do not change a setting from default unless you have a good reason too.
In life, gurus who are persistent -who make a mistake sometimes come up with a gem. People who have no idea what they are doing, and then make a mistake rarely come up with a gem
Use the methodology. It works well and this is reflected in my trading account balance - which is why we are all here.

So lets summarize some of the methodology
MAKE A BENCHMARK of 50,000 systems with 250 or 300 in Favorites A, change ONE thing at a time and see how it changes the benchmark
If you fail to do this, you are playing pin the tail on the donkey with your eyes closed.

We need to choose a market. Some markets are easy for GSB, some near impossible - but that can change as GSB architecture is expanded, or we crack the unique characteristic of each market.
Some markets are good, but not always hot to trade. CL later this year was just fair, but I think im JUST at new highs again on my first system since april.
Other markets @US for example are extremely hard.
If your a new user and your working on a market no one else has, your more likely to be in for a very slow journey - though there are many markets I suspect will work great with GSB.
If your a new user on a not well known market and low amount of CPU power, it will take you near forever to make progress.
It takes me a long time to learn a market, but the time decreases significantly each time. This is due to what we have learnt, GSB automation, and the massive power of GSB users working together.
GSB has community and collective knowledge. Being a lone ranger is not likely to work well.

So session time. Trade where the volume and range are. see video market validation1
https://trademaid.info/gsbhelp/Videos.html
So if you say, im going to try 24 hour data, you have broken the methodology, your results will be crap, and you have wasted a lot of time.
The exception is ES that can enter about 1715 and exit 1500, long only (not much work done on this, but I am live trading one non GSB system going this written by myself)

Overall day trading is best, and my preference and where my expertise is. If you do swing trading, likely you trade long only

settings, im going in order
Operators
use * only. + is almost as good, but needs other settings changed. To use +/- at once will give some different systems, but overall will be significantly worse and slow to build
Entry modes. Each market is unique. Never mix entry modes at the same time. IE use only 1. The exception is cross entry types which are so close to each other, its less critical

More tomorrow on this...
Normalization modes.
Beta users have access to 4 mode modes.
Highestlowest3 and centerscaling3 should never be used.
They were designed to fix MC or TS bug where highest / lowest gave incorrect results.
Unless the start date of gsb and ts/mc code is the same, results wont match.
If you change start date of ts chart, results will change. I likely will revoke these 2 normalization modes.
Centerscaling2 I think is slightly better than HighestLowest2
Weights mode. The short story is leave as is. Should you use + operand etc, might need a tweak to improve things.

SECONDARY FILTER.
This is critical to get best results.
If you don't know what your doing leave on AUTO.
indices need closedminusclosebpv (not normalized) and most other markets need GA (normally with closedminusClosebpv)
disable secondary filter for tick data or daily data. dont disable for intra day data
Today's build has a variety of countertrend non normalized SF

I suspect countertrend not normalized might work ok on ES, but have not tested yet. New GSB users best not experiment here.

meldinman - 16-12-2020 at 06:12 PM

Thanks for this write up, I think many people get confused and aren't properly using GSB due to sometimes unclear or conflicting information from various places (documentation, youtube channel, forum etc) It's easy to get tripped up when trying to make sure all settings are properly set when one is trying to build systems for a new market.

While automation has definitely helped ensure consistency of settings inputs, I think what may be helpful is a macro folder with settings for each market that has been developed so far with GSB. This would allow users to easily reset to the default settings for their market before they try to tweak any settings if they ever stray of course. Due to the various settings its very easy to miss something in a build on one of the settings option or become confused with which parameters to apply for correct methodology on that given market.
I know some of the markets already have a macro setting but I think a folder with one macro or each market may be useful?

admin - 16-12-2020 at 06:26 PM