|
Re: New Zorro version 1.83
[Re: jcl]
#472794
05/23/18 16:53
05/23/18 16:53
|
Joined: Apr 2018
Posts: 16
claudio
Newbie
|
Newbie
Joined: Apr 2018
Posts: 16
|
I'm tryng the MT5 bridge and this is the result I think is wrong. Note that if I buy 0.1 lot of EURUSD with my real account I will get 1000€ (about 1170$) with a margin cost of 100€ (Leverage 10).
This what I get from code added to TradeTest.c:
printf("n Price %.5f",InitialPrice); printf("n Spread %.5f",Spread); printf("n RollLong %.5f",RollLong); printf("n RollShort %.5f",RollShort); printf("n PIP %.5f",PIP); printf("n PIPCost %.5f",PIPCost); printf("n MarginCost %.5f",MarginCost); printf("n Leverage %.5f",Leverage); printf("n LotAmount %.5f",LotAmount); printf("n Commission %.5f",Commission);
TradeTest Login 0 MT4/5 (demo).. !Trading Point Of Financial Instruments UK LTD connected !Real account not supported at UTC 05-23 16:25 Bal 0.0 Equ 25.00 Mrg 0.0 Pos 53.00 Error 063: marketVol not available Vol 0.0 Spr 0.00017 V 1.831 on Wed 18-05-23 18:26:18
Trade: TradeTest EUR/USD 2018-05-23 Log Params EUR/USD Price 1.13795 Spread 0.00017 RollLong -11.01000 RollShort 4.69000 PIP 0.00010 PIPCost 0.00855 MarginCost 10.00000 Leverage 99.11544 LotAmount 100.00000 Commission 0.60000 Session end at 16:26 Logout.. ok
This what I get from MQL5 added in Zorro.mq5
Print("Asset ",Asset); Print("LotFactor ",LotFactor); Print("Factor ",Factor); Print("MODE_POINT ",MarketInfo(Asset,MODE_POINT)); Print("MODE_TICKVALUE ",MarketInfo(Asset,MODE_TICKVALUE)); Print("MODE_LOTSIZE ",MarketInfo(Asset,MODE_LOTSIZE)); Print("MODE_MARGINREQUIRED ",MarketInfo(Asset,MODE_MARGINREQUIRED)); Print("MODE_SWAPLONG ",MarketInfo(Asset,MODE_SWAPLONG)); Print("MODE_SWAPSHORT ",MarketInfo(Asset,MODE_SWAPSHORT)); Print("MODE_SWAPTYPE ",MarketInfo(Asset,MODE_SWAPTYPE)); Print("PERIOD_M15 ",iVolume(Asset,PERIOD_M15,0)); Print("ACCOUNT_LEVERAGE ",AccountInfoInteger(ACCOUNT_LEVERAGE)); Asset EURUSDmicro LotFactor 0.1 Factor 10.0 MODE_POINT 1e-05 MODE_TICKVALUE 0.008547227706693335 MODE_LOTSIZE 1000.0 MODE_MARGINREQUIRED 0.0 MODE_SWAPLONG -11.01 MODE_SWAPSHORT 4.69 MODE_SWAPTYPE 1.0 PERIOD_M15 2767 ACCOUNT_LEVERAGE 10
|
|
|
Re: New Zorro version 1.83
[Re: jcl]
#472940
05/31/18 22:20
05/31/18 22:20
|
Joined: Aug 2017
Posts: 4 Kent
jtu
Guest
|
Guest
Joined: Aug 2017
Posts: 4
Kent
|
Hello, I installed latest Zorro_1832.exe and tried Test for Z12. There are many Error 055 such as GBP/USD history missing (2018 8192 HistoryGBPUSD_2018.t6). Could you advise where to download XXXXXX_2018.t6 files? This zip file http://server.conitec.net/down/history_2016.zip does not have 2018 price data. Thanks
|
|
|
Re: New Zorro version 1.83
[Re: jtu]
#473003
06/07/18 09:10
06/07/18 09:10
|
Joined: Feb 2018
Posts: 68
tomaslolo
Junior Member
|
Junior Member
Joined: Feb 2018
Posts: 68
|
When using Deeplearn.c script if you modify BarPeriod and click Train, then you get Error 030: Check lookback, settings, asset order That doesn´t happen in 1.74.8 version.
|
|
|
Re: New Zorro version 1.83
[Re: Seymour]
#473034
06/10/18 21:06
06/10/18 21:06
|
Joined: Dec 2017
Posts: 14
vince
Newbie
|
Newbie
Joined: Dec 2017
Posts: 14
|
Forget my comment about the Z12.fac in Zorro 1.83.2. The problem was my AssetsFix.csv. More precisely: The RollLong and RollShort values for XAG/USD were off by a factor of 50! That wonders me because my assets file was created by Zorro after a test with connection to the MT4 bridge. It seems the Assets.csv generator in Zorro doesn't respect that according to http://manual.zorro-project.com/afaq.htm the rollover fee should be per contract! LotAmount is 50 for XAG/USD in my case. If I manually divide RollLong and RollShort by 50 and write that into the AssetsFix.csv then it is fine.
|
|
|
Re: New Zorro version 1.83
[Re: 3dgamelight]
#473038
06/11/18 04:12
06/11/18 04:12
|
Joined: Feb 2017
Posts: 369
Dalla
Senior Member
|
Senior Member
Joined: Feb 2017
Posts: 369
|
There is a lot of bugs in Zorro. If you have found a lot of bugs, this would be a good place to write them down so that the team can adress them.
|
|
|
Re: New Zorro version 1.83
[Re: Dalla]
#473057
06/11/18 18:37
06/11/18 18:37
|
Joined: Feb 2017
Posts: 1,718 Chicago
AndrewAMD
Serious User
|
Serious User
Joined: Feb 2017
Posts: 1,718
Chicago
|
Was there a release? I ask because: 1. This page says, "Zorro version 1.74.8 was released and is available on the download page." 2. This thread is no longer stickied.
|
|
|
Re: New Zorro version 1.83
[Re: jcl]
#473096
06/13/18 04:44
06/13/18 04:44
|
Joined: Jun 2018
Posts: 1 WA
drack
Guest
|
Guest
Joined: Jun 2018
Posts: 1
WA
|
Per http://zorro-project.com/manual/en/bittrex.htm "An asset list AssetsBittrex.csv with about 100 main cryto currencies is included" there is no assetbittrex.csv list in the download. It would be nice to have it since this is my first time using Zorro or an algo trader. If someone could Give me the format for the AssetsBittrex.csv compared to Bittrex's https://bittrex.com/api/v1.1/public/getcurrencies so I can try and sort this out that would be swell. [{"Currency":"BTC","CurrencyLong":"Bitcoin","MinConfirmation":2,"TxFee":0.00050000,"IsActive":true,"CoinType":"BITCOIN","BaseAddress":"1N52wHoVR79PMDishab2XmRHsbekCdGquK","Notice":null},
Last edited by drack; 06/13/18 04:47.
|
|
|
Re: New Zorro version 1.83
[Re: jcl]
#473113
06/14/18 08:23
06/14/18 08:23
|
Joined: Feb 2015
Posts: 652 Milano, Italy
MatPed
User
|
User
Joined: Feb 2015
Posts: 652
Milano, Italy
|
Hi, Zorro 1832 never finish training, in multicores, standard Workshop6. No issue with previous or later version. You can daownload the image from here:
https://biteu-my.sharepoint.com/:i:/g/personal/matteo_pedroni_bit-eu_com/EU5ihFE08V9GvZPqyFBeqPMBhOaQAiLOXvT6hwI4_govsQ?e=hmJamq
Ciao
|
|
|
Re: New Zorro version 1.83
[Re: jcl]
#473507
07/16/18 21:23
07/16/18 21:23
|
Joined: Jun 2013
Posts: 1,609
DdlV
Serious User
|
Serious User
Joined: Jun 2013
Posts: 1,609
|
Hi jcl. 2 things about 1.83: 1) I see above that it's released, however I didn't receive the normal release announcement email - was it sent? 2) I have downloaded and installed 1.832 (via Zorro_Setup) and the latest 2010 and 2016 History files. All Z Systems Test to the results in the manual, EXCEPT Z3 - it only Tests to 33% AR, not ~90%. It seems something like this happened once before, but I can't find that thread. If the Zorro version and History files are correct per the above, and all the other Systems Test fine, where else should I look to fix Z3 Test? FYI, here is the Z3.txt file I get:
WFA Test Z3 (TICKS)
Simulated account AssetsFix
Bar period 24 hours (avg 2027 min)
Simulation period 2010-05-24..2018-05-12 (2067 bars)
Test period 2012-07-23..2018-05-12 (1504 bars)
Lookback period 100 bars (20 weeks)
WFO test cycles 8 x 188 bars (38 weeks)
Training cycles 9 x 563 bars (116 weeks)
Montecarlo cycles 200
Simulation mode Realistic ticks (slippage 5.0 sec)
Gross win/loss 32111$ / -23282$ (+4997p)
Average profit 1522$/year, 127$/month, 5.85$/day
Max drawdown -5970$ 67.6% (MAE -5970$ 67.6%)
Total down time 86% (TAE 68%)
Max down time 47 weeks from Nov 2014
Max open margin 266$
Max open risk 3036$
Trade volume 3788968$ (653101$/year)
Transaction costs -1393$ spr, -983$ slp, -2491$ rol
Capital required 4559$
Number of trades 245 (43/year, 1/week, 1/day)
Percent winning 47.3%
Max win/loss 2052$ / -812$
Avg trade profit 36.04$ 20.4p (+156.7p / -102.1p)
Avg trade slippage -4.01$ -2.3p (+0.5p / -4.8p)
Avg trade bars 6 (+8 / -5)
Max trade bars 37 (7 weeks)
Time in market 108%
Max open trades 4
Max loss streak 11 (uncorrelated 9)
Annual return 33%
Profit factor 1.38 (PRR 1.15)
Sharpe ratio 0.64
Kelly criterion 1.24
R2 coefficient 0.015
Ulcer index 21.2%
Confidence level AR DDMax Capital
10% 38% 5244 4037$
20% 34% 5905 4513$
30% 31% 6523 4957$
40% 28% 7102 5373$
50% 26% 7681 5789$
60% 24% 8524 6395$
70% 22% 9260 6925$
80% 20% 10472 7796$
90% 15% 13502 9975$
95% 13% 15586 11474$
100% 9% 22342 16332$
Portfolio analysis OptF ProF Win/Loss Wgt% Cycles
NAS100 avg .001 1.04 27/38 3.0 /\\//
SPX500 avg .000 0.82 26/32 -7.3 /\..//
US30 avg .042 2.08 39/34 58.8 ////////
XAG/USD avg .014 1.52 24/25 45.5 /XX///.
NAS100 .003 1.04 27/38 3.0 /\\//
NAS100:L .003 1.04 27/38 3.0 /\\//
NAS100:S .000 ---- 0/0 0.0 ........
SPX500 .000 0.82 26/32 -7.3 /\..//
SPX500:L .000 0.82 26/32 -7.3 /\..//
SPX500:S .000 ---- 0/0 0.0 ........
US30 .084 2.08 39/34 58.8 ////////
US30:L .084 2.08 39/34 58.8 ////////
US30:S .000 ---- 0/0 0.0 ........
XAG/USD .014 1.52 24/25 45.5 /XX///.
XAG/USD:L .007 1.31 8/13 14.6 //./..
XAG/USD:S .021 1.75 16/12 30.9 //\///.
Thanks.
|
|
|
Re: New Zorro version 1.83
[Re: DdlV]
#473535
07/20/18 13:16
07/20/18 13:16
|
Joined: Jul 2000
Posts: 27,977 Frankfurt
jcl
OP
Chief Engineer
|
OP
Chief Engineer
Joined: Jul 2000
Posts: 27,977
Frankfurt
|
Gross win/loss 24599$ / -13393$ (+6458p)
Average profit 1932$/year, 161$/month, 7.43$/day
Max drawdown -2435$ 21.7% (MAE -3559$ 31.8%)
Total down time 72% (TAE 54%)
Max down time 47 weeks from Nov 2014
Max open margin 266$
Max open risk 2954$
Trade volume 2400653$ (413798$/year)
Transaction costs -943$ spr, -639$ slp, -1803$ rol
Capital required 2017$
Number of trades 157 (28/year, 1/week, 1/day)
Percent winning 53.5%
Max win/loss 2052$ / -812$
Avg trade profit 71.38$ 41.1p (+168.8p / -105.7p)
Avg trade slippage -4.07$ -2.3p (+0.6p / -5.7p)
Avg trade bars 7 (+8 / -5)
Max trade bars 37 (7 weeks)
Time in market 76%
Max open trades 4
Max loss streak 7 (uncorrelated 7)
Annual return 96%
Profit factor 1.84 (PRR 1.46)
Sharpe ratio 1.00
Kelly criterion 1.05
R2 coefficient 0.744
Ulcer index 6.5%
|
|
|
Re: New Zorro version 1.83
[Re: jcl]
#473540
07/20/18 16:40
07/20/18 16:40
|
Joined: Jun 2013
Posts: 1,609
DdlV
Serious User
|
Serious User
Joined: Jun 2013
Posts: 1,609
|
Thanks jcl. All Assets are giving different results. To account for that, it seems the likely possibilities are different AssetsFix.csv and/or History files. The AssetsFix.csv that was installed by 1.832 is dated Feb 1 and contains:
Name,Price,Spread,RollLong,RollShort,PIP,PIPCost,MarginCost,Leverage,LotAmount,Commission,Symbol
AUD/USD,0.77311,0.00005,0.24,-0.51,0.0001,0.0871,9,0,1000,0.6,
EUR/CHF,1.07962,0.00004,0.08,-0.26,0.0001,0.09263,75,0,1000,0.8,
EUR/JPY,123.359,0.006,-0.3,0.09,0.01,0.081066,10,0,1000,0.8,
EUR/USD,1.13795,0.00005,-0.02,0.01,0.0001,0.0871,10,0,1000,0.6,
GBP/USD,1.51364,0.00003,0.05,-0.12,0.0001,0.0871,14,0,1000,0.6,
GER30,10884,1,-0.1,0,1,0.1,6,0,0.1,0,GER30
NAS100,4185,1,-0.1,0,1,0.0871,2,0,0.1,0,NAS100
NZD/USD,0.67752,0.00007,0.6,-1.32,0.0001,0.088713,7,0,1000,0.8,
SPX500,2032.4,0.5,-0.5,0,0.1,0.0871,8,0,1,0,
UK100,6869,2,-0.1,0,1,0.13183,6,0,0.1,0,
US30,17520,2,-0.5,0,1,0.0871,6,0,0.1,0,
USD/CAD,1.24466,0.0001,-0.21,0.09,0.0001,0.06998,10,0,1000,0.8,
USD/CHF,0.92203,0.00022,0.15,-0.58,0.0001,0.09446,50,0,1000,0.6,
USD/JPY,117.489,0.004,0.02,-0.04,0.01,0.07413,10,0,1000,0.6,
XAG/USD,17.235,0.046,-0.0009,0,0.01,0.43547,5,0,50,0,
XAU/USD,1260.82,0.47,-0.0124,0,0.01,0.00871,11,0,1,0,
Does it match yours? Thanks.
|
|
|
Re: New Zorro version 1.83
[Re: jcl]
#473685
08/04/18 04:44
08/04/18 04:44
|
Joined: Jun 2013
Posts: 1,609
DdlV
Serious User
|
Serious User
Joined: Jun 2013
Posts: 1,609
|
Thanks for clarifying. In the meantime, while waiting impatiently for the next version, I still don't quite understand about the outlier/cluster, and how just adding more bars eliminates it. When you have a sec, would appreciate a little more detail. Thanks.
|
|
|
|