1MDAX/RSI/MACD

This topic has 32 replies, 9 voices, and was last updated 5 years ago by Jaykay.

Viewing 15 posts - 1 through 15 (of 33 total)

1 2 3 →

07/20/2018 at 12:37 PM #76010

Jaykay

I was backtesting a strategy on the 1 minute chart and came up with this very simple algorithm based on the RSI and MACD indicators.

Its a long only strategy with fixt SL and TP

When the RSI =<25. and the MACD line =<-10. it takes a long position with SL 30pt and TP50pt

postion size is a bit progressive at 10pp with an account of 10.000,-

It seems to work best in downtrends.

A backtest at 200.000 units or any suggestions r welcome.

Jaykay

Participant

Topics: 3

Replies: 18

Been thanked: 3 times

07/21/2018 at 7:44 PM #76432

GraHal

Looks good! Mean reverting strategy as you are buying when oversold?

Weird though … I get different results, still respectable for a 1 min TF, but I started and finished a week later than you and my Win % is 29% whereas your Win % is 57%??

Good Work!
GraHal

PS did you try similar for Sellshort when overbought??

1 user thanked author for this post.

Jaykay

GraHal

Participant

Topics: 204

Replies: 8399

Been thanked: 2041 times

07/21/2018 at 8:28 PM #76436

Jaykay

Thank GraHal for your response i did try it for selling short but it was a disaster.

Weird is that you have so many even trades because its not programmed to break even. it looks like it could be a good idea.

I have several versions running in demo and in event of trumps latest tweets

a SL of 35 seems to do better.

Jaykay

Participant

Topics: 3

Replies: 18

Been thanked: 3 times

07/21/2018 at 8:39 PM #76439

Vonasi

Jaykay wrote:

Weird is that you have so many even trades because its not programmed to break even.

Maybe you are having the same issue as I am with 0.00 abs perf trades. They show up as even trades.

https://www.prorealcode.com/topic/0-00-abs-perf-in-backtest-with-cumulating-positions/

Vonasi

Moderator

Topics: 192

Replies: 5088

Been thanked: 1810 times

07/21/2018 at 9:54 PM #76440

GraHal

Vonasi wrote:

Maybe you are having the same issue as I am with 0.00 abs perf trades. They show up as even trades.

Well I am yes as I noted on your Topic on 0.00 perf trades.

@JayKay run yours on 100k bars up to 20 Jul? I bet you get lots of even trades also as it is a glitch that has come to light in the past week since the 12 Jul 18 update!?

Hope they sort it soon, but as it’s the Demo Platform it’s low on PRT Priorities.

GraHal

Participant

Topics: 204

Replies: 8399

Been thanked: 2041 times

07/21/2018 at 10:13 PM #76441

Jaykay

@Grahal I have no glitch on demo or real time platform with various starting and ending dates.

Jaykay

Participant

Topics: 3

Replies: 18

Been thanked: 3 times

07/22/2018 at 5:05 AM #76442

Nobody

The problem with some systems such as this are no entries in strong uptrends where retraces are shallow . On the other side you get a lot of entries on downtrends due to opposite , maybe a trend and momentum filter will help . Will have a shot at this over the next day or so . Finding areas where an algo doesnt perform is the key to improving . The equity curve to trend relationship reveals a lot . There is probably room to run 2 algos at once to solve some of the problems preseneted by this system .

Maybe take the standard entry in shallow trends and possibly relax the entry standards in a high momentum uptrend . Turn of system in steep downtrends . Im guessing you will get less trades at a higher WR making for a way smoother curve and potential for slightly larger trade size . A smoother curve makes way for a dynamic position sizing to get that more J curve thing going . Just a couple ideas fwiw

1 user thanked author for this post.

Jaykay

Nobody

Participant

Topics: 9

Replies: 190

Been thanked: 59 times

07/22/2018 at 8:39 AM #76443

GraHal

Jaykay wrote:

I have no glitch on demo or real time platform with various starting and ending dates.

Please post screen shot of win % up to 20 Jul over 100k bars as I cannot believe that only me, Vonasi and Grubby are the only 3 users getting the 0.00 abs perf trades??

GraHal

Participant

Topics: 204

Replies: 8399

Been thanked: 2041 times

07/22/2018 at 10:26 AM #76445

Jaykay

GraHal wrote:

Jaykay wrote:

I have no glitch on demo or real time platform with various starting and ending dates.

Please post screen shot of win % up to 20 Jul over 100k bars as I cannot believe that only me, Vonasi and Grubby are the only 3 users getting the 0.00 abs perf trades??

Jaykay

Participant

Topics: 3

Replies: 18

Been thanked: 3 times

07/22/2018 at 10:48 AM #76447

GraHal

Ah so no even trades then … well weird!

So what version Demo are you showing JayKay under Help, About??

GraHal

Participant

Topics: 204

Replies: 8399

Been thanked: 2041 times

07/22/2018 at 11:10 AM #76448

GraHal

I left Longs as per JayKay version, but I added Shorts … see top equity curve / V3.0.

I retained the same strategy … Mean Reverting using RSI and MACD, but I added Multi-TF to the Short side.

Results attached showing Long only, Short only and Longs and Shorts … but bear in mind the Phantom Even Trades when judging the win%.

Please feel free to improve more and post on this Topic.

// Definition of code parameters
DEFPARAM CumulateOrders = False // Cumulating positions deactivated

noEntryBeforeTime = 090000
timeEnterBefore = time >= noEntryBeforeTime

noEntryAfterTime = 173000
timeEnterAfter = time < noEntryAfterTime

// Conditions to enter long positions
indicator1 = RSI[14](close)
c1 = (indicator1 <= 25)
indicator2 = MACDline[12,26,9](close)
c2 = (indicator2 <= -10)

IF (c1 AND c2) AND timeEnterBefore AND timeEnterAfter THEN
BUY 1 CONTRACT AT MARKET
ENDIF

// Stops and targets
If LongonMarket then
SET STOP pLOSS 30
SET TARGET pPROFIT 50
endif


// GO SHORT
Timeframe (5mn)
indicator3 = RSI[14](close)
c3 = (indicator3 >= 70)
Timeframe (1mn)
indicator4 = MACDline[12,26,9](close)
c4 = (indicator4 >= 5)

IF (c3 AND c4) THEN
SellShort 1 CONTRACT AT MARKET
ENDIF

If ShortonMarket Then
SET STOP pLOSS 70 //30
SET TARGET pPROFIT 40 //50
Endif

// Definition of code parameters

DEFPARAM CumulateOrders = False // Cumulating positions deactivated

noEntryBeforeTime = 090000

timeEnterBefore = time >= noEntryBeforeTime

noEntryAfterTime = 173000

timeEnterAfter = time < noEntryAfterTime

// Conditions to enter long positions

indicator1 = RSI[14](close)

c1 = (indicator1 <= 25)

indicator2 = MACDline[12,26,9](close)

c2 = (indicator2 <= -10)

IF (c1 AND c2) AND timeEnterBefore AND timeEnterAfter THEN

BUY 1 CONTRACT AT MARKET

ENDIF

// Stops and targets

If LongonMarket then

SET STOP pLOSS 30

SET TARGET pPROFIT 50

endif

// GO SHORT

Timeframe (5mn)

indicator3 = RSI[14](close)

c3 = (indicator3 >= 70)

Timeframe (1mn)

indicator4 = MACDline[12,26,9](close)

c4 = (indicator4 >= 5)

IF (c3 AND c4) THEN

SellShort 1 CONTRACT AT MARKET

ENDIF

If ShortonMarket Then

SET STOP pLOSS 70 //30

SET TARGET pPROFIT 40 //50

Endif

2 users thanked author for this post.

Berra,

Jaykay

GraHal

Participant

Topics: 204

Replies: 8399

Been thanked: 2041 times

07/22/2018 at 12:43 PM #76451

Nobody

Jaykay wrote:

I was backtesting a strategy on the 1 minute chart and came up with this very simple algorithm based on the RSI and MACD indicators.

Its a long only strategy with fixt SL and TP

When the RSI =<25. and the MACD line =<-10. it takes a long position with SL 30pt and TP50pt

postion size is a bit progressive at 10pp with an account of 10.000,-

It seems to work best in downtrends.

A backtest at 200.000 units or any suggestions r welcome.

Little confused as to how your best trade is >1000 when TP is 50p which at 10 times is 500 . What am i missing ?

Nobody

Participant

Topics: 9

Replies: 190

Been thanked: 59 times

07/22/2018 at 1:08 PM #76452

Jaykay

Brisvegas wrote:

Little confused as to how your best trade is >1000 when TP is 50p which at 10 times is 500 . What am i missing ?

Thanx @Brisvagas for your good suggestions. I wil definitely try to implicate them

about the big winner, i’ve notices the same

but it doesn’t show at my latest backtest.

since the strategy can go overnight could it be positive slippage???

Jaykay

Participant

Topics: 3

Replies: 18

Been thanked: 3 times

07/22/2018 at 1:11 PM #76453

GraHal

Only thing I can think of is that trade was already in profit maybe 45 points ish (so at a lot size of 10 = 450) then next morning at open there was a 55 point gap in the trades favour so then the trade closed at 1000 gain (100 x 10 lot size) ??

Or something similar ?? 🙂

GraHal

Participant

Topics: 204

Replies: 8399

Been thanked: 2041 times

07/22/2018 at 1:13 PM #76454

Jaykay

Wow, @GraHal Thanks for your contribution to the strategy!! love the MTF

Wil get into it over the next days!!

THANX!!

1 user thanked author for this post.

GraHal

Jaykay

Participant

Topics: 3

Replies: 18

Been thanked: 3 times

Author

Posts

Viewing 15 posts - 1 through 15 (of 33 total)

1 2 3 →

Find exclusive trading pro-tools on