Confirmation of Trend using Neural Networks (by kind permission of Leo)

GraHal · Fri, 24 Aug 2018 13:45:40 +0000

TOPIC SPLITTED FROM “Neural Networks programming” discussion: https://www.prorealcode.com/topic/neural-networks-programming-with-prorealtime/ to discuss the strategy made by GraHal about Leo’s classifier code snippet. Leo wrote: which function do you use for the output neurons? Nicolas wrote: how did GraHal use the code to trigger trades in this case? I’m far from expert at coding, so I couldn’t understand your code fully, but I do like experimenting so … (rightly or wrongly?) I did below to trigger the trades! 🙂 System is doing good so far on Fwd Test … current trade at £128 profit and past trades results attached IF classifier=1 and close > Average[60](close) then Buy 2 Contracts at Market Edit / PS Bear in mind results are on 5 min TF, so maybe the good results so far are a fluke and a misuse of the code intention?? Strategy using the neural network made by Leo starts from this post: https://www.prorealcode.com/topic/long-only-trading-with-trend-confirmation-on-5-minutes-timeframe/page/3/#post-79578

This topic has 37 replies, 8 voices, and was last updated 6 years ago by GraHal.

Tagged: data mining, machine learning, neural network

Viewing 15 posts - 16 through 30 (of 38 total)

← 1 2 3 →

08/28/2018 at 7:03 AM #79212

raphaelopilski

this would be nice for the dax. Because of the Margin. It´s the half of Dow. Anyone tried?

raphaelopilski

Participant

Topics: 16

Replies: 73

Been thanked: 4 times

08/28/2018 at 8:57 AM #79220

GraHal

Mmm thought it was a bit too good to be true! Just a bit though! 🙂

I have done a pessimistic calculation / check of results and £300 – £400 of profit is due to the spike we have had to endure at 21:15 on the Dow for the last 5 nights or so.

4 trades have been affected at 21:15, 1 lost around 350 due to the spike and 3 gained 200 due to the spike!

The 4 spiked tradess may have gone on to be gainers anyway if the spike hadn’t triggered the SL and TP. I guess I could work this out, but the coffee hasn’t worked through yet! 🙂

Results are still good though as v1.0 is currently trading at a genuine around £600 profit over 6 days (950 – 350 due to IG spikes! ).

I will put a time filter on the strategy when I get chance.

GraHal

Participant

Topics: 214

Replies: 8830

Been thanked: 2201 times

08/29/2018 at 11:44 AM #79288

GraHal

Vonasi proposed so here it is … SPTRD on 1 Min TF 100k bars, spread = 4

DEFPARAM CUMULATEORDERS = False
//Variables:
candlesback=5
ProfitRiskRatio=2
spread=1.5
 
 
myATR=average[20](range)+std[20](range)
ExtraStopLoss=MyATR
//ExtraStopLoss=3*spread*pipsize
 
//for long trades
classifierlong=0
FOR scanL=1 to candlesback DO
IF classifierlong[scanL]=1 then
BREAK
ENDIF
LongTradeLength=ProfitRiskRatio*(close[scanL]-(low[scanL]-ExtraStopLoss[scanL]))
IF close[scanL]+LongTradeLength < high-spread*pipsize then
IF lowest[scanL+1](low) > low[scanL]-ExtraStopLoss[scanL]+spread*pipsize then
classifierlong=1
//candleentrylong=barindex-scanL
BREAK
ENDIF
ENDIF
NEXT
IF classifierlong=1 and Close > Average[120](close)then
//DRAWSEGMENT(candleentrylong,close[barindex-candleentrylong],barindex,close[barindex-candleentrylong]+LongTradeLength) COLOURED(0,150,0)
//DRAWELLIPSE(candleentrylong-1,low[barindex-candleentrylong]-ExtraStopLoss,barindex+1,high+ExtraStopLoss) COLOURED(0,150,0)
Buy 2 Contract at Market
ENDIF

//for short trades
classifiershort=0
FOR scanS=1 to candlesback DO
IF classifiershort[scanS]=1 then
BREAK
ENDIF
ShortTradeLength=ProfitRiskRatio*((high[scanS]-close[scanS])+ExtraStopLoss[scanS])
IF close[scanS]-ShortTradeLength > low+spread*pipsize then
IF highest[scanS+1](high) < high[scanS]+ExtraStopLoss[scanS]-spread*pipsize then
classifiershort=1
//candleentryshort=barindex-scanS
BREAK
ENDIF
ENDIF
NEXT
 
IF classifiershort=1 and Close < Average[120](close) then
//DRAWSEGMENT(candleentryshort,close[barindex-candleentryshort],barindex,close[barindex-candleentryshort]-ShortTradeLength) COLOURED(150,0,0)
//DRAWELLIPSE(candleentryshort-1,high[barindex-candleentryshort]+ExtraStopLoss,barindex+1,low-ExtraStopLoss) COLOURED(150,0,0)
SellShort 2 Contract at Market
ENDIF
 
SET TARGET PROFIT 70
SET STOP PLOSS 35

DEFPARAM CUMULATEORDERS = False

//Variables:

candlesback=5

ProfitRiskRatio=2

spread=1.5

myATR=average[20](range)+std[20](range)

ExtraStopLoss=MyATR

//ExtraStopLoss=3*spread*pipsize

//for long trades

classifierlong=0

FOR scanL=1 to candlesback DO

IF classifierlong[scanL]=1 then

BREAK

ENDIF

LongTradeLength=ProfitRiskRatio*(close[scanL]-(low[scanL]-ExtraStopLoss[scanL]))

IF close[scanL]+LongTradeLength < high-spread*pipsize then

IF lowest[scanL+1](low) > low[scanL]-ExtraStopLoss[scanL]+spread*pipsize then

classifierlong=1

//candleentrylong=barindex-scanL

BREAK

ENDIF

IF classifierlong=1 and Close > Average[120](close)then

//DRAWSEGMENT(candleentrylong,close[barindex-candleentrylong],barindex,close[barindex-candleentrylong]+LongTradeLength) COLOURED(0,150,0)

//DRAWELLIPSE(candleentrylong-1,low[barindex-candleentrylong]-ExtraStopLoss,barindex+1,high+ExtraStopLoss) COLOURED(0,150,0)

Buy 2 Contract at Market

ENDIF

//for short trades

classifiershort=0

FOR scanS=1 to candlesback DO

IF classifiershort[scanS]=1 then

BREAK

ENDIF

ShortTradeLength=ProfitRiskRatio*((high[scanS]-close[scanS])+ExtraStopLoss[scanS])

IF close[scanS]-ShortTradeLength > low+spread*pipsize then

IF highest[scanS+1](high) < high[scanS]+ExtraStopLoss[scanS]-spread*pipsize then

classifiershort=1

//candleentryshort=barindex-scanS

BREAK

ENDIF

IF classifiershort=1 and Close < Average[120](close) then

//DRAWSEGMENT(candleentryshort,close[barindex-candleentryshort],barindex,close[barindex-candleentryshort]-ShortTradeLength) COLOURED(150,0,0)

//DRAWELLIPSE(candleentryshort-1,high[barindex-candleentryshort]+ExtraStopLoss,barindex+1,low-ExtraStopLoss) COLOURED(150,0,0)

SellShort 2 Contract at Market

ENDIF

SET TARGET PROFIT 70

SET STOP PLOSS 35

GraHal

Participant

Topics: 214

Replies: 8830

Been thanked: 2201 times

08/29/2018 at 8:09 PM #79311

Vonasi

GraHal wrote:

Vonasi proposed so here it is … SPTRD on 1 Min TF 100k bars, spread = 4

Why 4 pips? The SP500 spread is never that big.

Vonasi

Moderator

Topics: 192

Replies: 5088

Been thanked: 1811 times

08/29/2018 at 8:18 PM #79312

4everTrading

Hey Everyone!!

Such a nice work!! Amazing results from the beginning, Congrats guys!!!

I would like my little idea to improve the results. Months ago I added an indicator to the library. I think this indicator can prevent fake entries, basiclly I added two more conditions to the original code:

AverageVolumeBuy > AverageVolumeSell and AverageVolumeBuy > 50

1	AverageVolumeBuy > AverageVolumeSell and AverageVolumeBuy > 50

It is only a little bit improve, and please guys please to share your opinion. I hope you like.

Saludos!!

1 user thanked author for this post.

Nicolas

4everTrading

Participant

Topics: 1

Replies: 19

Been thanked: 3 times

08/29/2018 at 10:35 PM #79333

GraHal

Vonasi wrote:

Why 4 pips? The SP500 spread is never that big.

I forgot, I forgot okay!? We all make mistakes!? 🙂

Also Lot size = 2 on my results above.
S&P500 is min Lot size 5, so more mistakes on my part! 🙁

Xusto wrote:

I hope you like.

Yeah looks good, thanks! The more the merrier! 🙂

GraHal

Participant

Topics: 214

Replies: 8830

Been thanked: 2201 times

09/02/2018 at 8:37 AM #79542

Stefanb

GraHal

200k on your SP500 spread=2.

1 user thanked author for this post.

GraHal

Stefanb

Participant

Topics: 15

Replies: 74

Been thanked: 22 times

09/02/2018 at 8:47 AM #79544

Stefanb

I did a 200k test on your code.

Stefanb

Participant

Topics: 15

Replies: 74

Been thanked: 22 times

09/02/2018 at 8:52 AM #79548

GraHal

@Stefanb @Xusto @anybody … got any tweaks / fixes for that big drawdown shown on Stefanb screenshot over 200k bars above??

Please post your tweaked / fixed code on here with screen shot of equity curve and performance results.

GraHal

Participant

Topics: 214

Replies: 8830

Been thanked: 2201 times

09/02/2018 at 8:54 AM #79549

Nicolas

Equity curve stagnation and lose come from ranging market. This seems normal for a strategy that detects market movements and bets on their continuity 🙂

Nicolas

Keymaster

Topics: 81

Replies: 20863

Been thanked: 4774 times

09/02/2018 at 8:57 AM #79550

GraHal

Well we need code to keep out of the market when ranging?

GraHal

Participant

Topics: 214

Replies: 8830

Been thanked: 2201 times

09/02/2018 at 9:01 AM #79551

GraHal

Stefanb wrote:

X I did a 200k test on your code.

Above is for Xusto version with the volume tweak in post 79312.

( @stefanb in case you don’t know and for others … if you select part of the text of a post and THEN click on Quote on the right above the post then the selected text displays as a quote … very useful).

GraHal

Participant

Topics: 214

Replies: 8830

Been thanked: 2201 times

09/02/2018 at 10:10 AM #79557

Stefanb

Here is a code that I will start live soon.

Have run it om demo for a while and it seems to be working ok.

Spread=1

Op 75%

DEFPARAM CumulateOrders = False
DEFPARAM PRELOADBARS = 200

indicator1 = SuperTrend[1,8]
c1 = (close <= indicator1)
indicator4 = SuperTrend[1,4]
c4 = (close <= indicator4)
c2 =  DIplus[6](close) >  DIminus[3](close)

HL = 14 // Hour of Long Entry
HLS = 12 // Hour of Long Exit
HS = 7 // Hour of Short Entry
HSS = 15 // Hour of Short Exit
LATR = 9// Number of ATRs for Stop Loss
PATR = 21 // Number of ATRs for Take Profit
ATR = 18 // ATR period

IF Hour=HL and c1 THEN
BUY 1 CONTRACTS AT MARKET
ENDIF

If LongOnMarket AND Hour=HLS THEN//
SELL AT MARKET
ENDIF

IF Hour=HS and c2 and c4 THEN
SELLSHORT 1 CONTRACTS AT MARKET
ENDIF

IF ShortOnMarket AND Hour=HSS THEN
EXITSHORT AT MARKET
ENDIF

SET STOP LOSS LATR*AverageTrueRange[ATR](close)
SET TARGET PROFIT PATR*AverageTrueRange[ATR](close)

IF Hour=HS and c2 and c4 THEN
SELLSHORT 1 CONTRACTS AT MARKET
ENDIF

IF ShortOnMarket AND Hour=HSS THEN
EXITSHORT AT MARKET
ENDIF

SET STOP LOSS LATR*AverageTrueRange[ATR](close)
SET TARGET PROFIT PATR*AverageTrueRange[ATR](close)

DEFPARAM CumulateOrders = False

DEFPARAM PRELOADBARS = 200

indicator1 = SuperTrend[1,8]

c1 = (close <= indicator1)

indicator4 = SuperTrend[1,4]

c4 = (close <= indicator4)

c2 = DIplus[6](close) > DIminus[3](close)

HL = 14 // Hour of Long Entry

HLS = 12 // Hour of Long Exit

HS = 7 // Hour of Short Entry

HSS = 15 // Hour of Short Exit

LATR = 9// Number of ATRs for Stop Loss

PATR = 21 // Number of ATRs for Take Profit

ATR = 18 // ATR period

IF Hour=HL and c1 THEN

BUY 1 CONTRACTS AT MARKET

ENDIF

If LongOnMarket AND Hour=HLS THEN//

SELL AT MARKET

ENDIF

IF Hour=HS and c2 and c4 THEN

SELLSHORT 1 CONTRACTS AT MARKET

ENDIF

IF ShortOnMarket AND Hour=HSS THEN

EXITSHORT AT MARKET

ENDIF

SET STOP LOSS LATR*AverageTrueRange[ATR](close)

SET TARGET PROFIT PATR*AverageTrueRange[ATR](close)

IF Hour=HS and c2 and c4 THEN

SELLSHORT 1 CONTRACTS AT MARKET

ENDIF

IF ShortOnMarket AND Hour=HSS THEN

EXITSHORT AT MARKET

ENDIF

SET STOP LOSS LATR*AverageTrueRange[ATR](close)

SET TARGET PROFIT PATR*AverageTrueRange[ATR](close)

2 users thanked author for this post.

4everTrading,

GraHal

Stefanb

Participant

Topics: 15

Replies: 74

Been thanked: 22 times

09/02/2018 at 10:33 AM #79559

robertogozzi

Lines 34-35 are duplicate and you may remove them.

Moreover, your SL & TP will change each new candle, while they should not while on market, I think you should use line 44 to write

If Not OnMarket then

1	If Not OnMarket then

and add line 47 with ENDIF

robertogozzi

Moderator

Topics: 69

Replies: 17188

Been thanked: 4064 times

09/02/2018 at 10:46 AM #79561

GraHal

Lines 37 to 46 are all duplicated?

I guess maybe finger trouble with copy and paste??

GraHal

Participant

Topics: 214

Replies: 8830

Been thanked: 2201 times

Author

Posts

Viewing 15 posts - 16 through 30 (of 38 total)

← 1 2 3 →

Find exclusive trading pro-tools on