intraday DAX strategy mini-1€, timeframe 5 min
Good afternoon,
This robot simply tries to buy or sell in the direction of the first 5 minutes of the market, ie if the first bar 5 minutes from the opening at 9am is positive, we buy, otherwise we sell.
With a spread of 1 point in the DAX mini 1 € and a time frame of 5 min.
In addition, it has an accumulator, in case the operation wins, the next day we will add 2 contracts, in case of losing the operation, we will divide between two contracts that we put the next day. Example:
Day 1: 1 contract -> We win; Day 2: 3 contracts-> We win; Day 3: 5 contracts-> We lost; Day 4: 3 (2.5) contracts-> We win; Day 5 = 5 contracts …
In addition, it has a manual multiplier to assume greater or lesser risk with the number of contracts.
There are also a maximum number of contracts.
IMPORTANT: For a faithful backtest, you must use PRT v10.3, in v.10.2 the results are NOT real.
A discussion thread is ongoing on the forum about the settings and feedback about the strategy, you can read, ask your questions and participate here: https://www.prorealcode.com/topic/5-min-mini1e-dax-spread-1/
//—————–
Buenas tardes,
Este robot trata simplemente de comprar o vender en la dirección de los primeros 5 minutos del mercado, es decir, que si la primera barra de 5 minutos desde la apertura a las 9h es positiva, compramos, en caso contrario, vendemos.
Con un spread de 1 punto en el DAX mini 1€ y un time frame de 5 min.
Además, tiene un acumulador, en el caso de que la operación resulte ganadora, al día siguiente sumaremos 2 contratos, en el caso de perder la operación, dividiremos entre dos los contratos que metemos al día siguiente. Ejemplo:
Día 1: 1 contrato -> Ganamos; Día 2: 3 contratos-> Ganamos; Día 3: 5 contratos-> Perdemos; Día 4: 3 (2,5) contratos-> Ganamos; Día 5= 5 contratos…
Además, tiene un multiplicador manual para asumir mayor o menor riesgo con la cantidad de contratos.
También hay puesto un máximo de contratos.
IMPORTANTE: Para un backtest fiel, hay que usar PRT v10.3, en la v.10.2 los resultados NO son reales.
// Definición de los parámetros del código
DEFPARAM CumulateOrders = false // Acumulación de posiciones desactivada
// La posición se cierra a las 17:29 p.m. si no toca ni stop ni take.
DEFPARAM FlatAfter =173000
once ordersize=1
// No se abren nuevas posiciones después de la vela que se cierra a las 09:06
HoraEntradaLimite = 090600
// El análisis de mercado empieza en la vela de 5 minutos que cierra a las 09:05
HoraInicio = 090500
// Órdenes máximas
if Ordersize>18 then
Ordersize=18
endif
// Riesgo, multiplicador de contratos.
n=6
// Condiciones para el analisis.
if Time >= HoraInicio and time <= HoraEntradaLimite then
// Condiciones de entrada de posiciones largas.
c1 = open < close-1
IF c1 THEN
IF PositionPerf(1) < 0 THEN
OrderSize = OrderSize/2//+1
if ordersize<1 then
ordersize=1
ENDIF
ELSIF PositionPerf(1) > 0 THEN
OrderSize =OrderSize+2
endif
// Si la primera barra del día de 5 min es positiva, compramos.
buy ordersize*n shares at market
endif
// Condiciones de entrada de posiciones cortas.
c2= open > close-1
IF c2 THEN
iF PositionPerf(1) < 0 THEN
OrderSize = OrderSize/2//+1
if ordersize<1 then
ordersize=1
ENDIF
ELSIF PositionPerf(1) > 0 THEN
OrderSize =OrderSize+2
ENDIF
// Si la primera barra del día de 5 min es negativa, vendemos.
sellshort ordersize*n shares at market
endif
endif
SET STOP ploss 5
SET TARGET pPROFIT 10
Comments
You must be logged in to post a comment.
Bonjour Nicolas,
comment se fait-il qu'avec cette stratégie stop loss à 5 pips, PRT mettant le stop à 17 pips ???????
sur 42 contrats,ça fait mal.y a t-il une explication ?les ordres étant passés en même temps(achat-vente stop limite)
Personnellement je ne comprends pas
All. These are the results with the version of the code I posted above. I know the code has changed but I have been asked for the results with a stop of 5.
Stop 5, Target 10: initial equity of EUR10k is too small. You have a max drawdown of EUR9063 and the strategy stops in Feb 2014. If you increase starting equity to EUR20k, strategy moves into profit from approx Q2 2014. It does really well but you have to be comfortable with the drawdown. Obviously returns aren't as high as you have doubled starting equity.
Stop 5, Target 12: Starting equity of 10k. Works OK. Drawdown of 6,014 (so more than when stop is 3). Makes more overall (+968% over the 4 years). Winning trades 35.62%, G/L 1.5, Loss of worst trade 1,651.
Overall, if you are just concerned with the total gain over the 4 years, stop 5 & target 12 is better. However I wasn't happy with the drawdown. The strategy "ticks along" through 2013 no matter what the settings (either small gain or, in most cases, loss). I optimised and walked forward from 2013 and I decided on Stop 3, Target 12. I was tempted with Stop 4, Target 10 (Gain 976%, % winning 36%, G/L 1.75) however had drawdown 5,834 and in particular had a worse year in 2013.
Hope this is of use.
Yes, I have noticed, has been because in backtest the openings and closures are perfect and live if there are fast movements are delayed, if you look at the price of entry in the backtest is different from the live entry. Plus today has been a day that has just gone up 5 points, this is scalping :( To solve this problem, I have modified the code. You have it in these last comments, instead of opening directly, it creates opening orders. I will create in the forum a topic.
with other accumulator:
// Definición de los parámetros del código
DEFPARAM CumulateOrders = false // Acumulación de posiciones desactivada
DEFPARAM FlatAfter =172900
once ordersize=1
HoraEntradaLimite = 090600
HoraInicio = 090500
if Ordersize>50 then
Ordersize=10
endif
if Time >= HoraInicio and time <= HoraEntradaLimite then
// Condiciones de entrada de posiciones cortas
c1 = open < close-2
if not onmarket then
IF c1 THEN
IF PositionPerf(1) < 0 THEN
OrderSize = OrderSize-1
if ordersize<1 then
ordersize=1
ENDIF
ELSIF PositionPerf(1) > 0 THEN
OrderSize =OrderSize+2
if ordersize<1 then
ordersize=1
ENDIF
endif
buy ordersize shares AT close[1]+2 stop
endif
c2= open > close-1
IF c2 THEN
iF PositionPerf(1) < 0 THEN
OrderSize = OrderSize-1
if ordersize<1 then
ordersize=1
ENDIF
ELSIF PositionPerf(1) > 0 THEN
OrderSize =OrderSize+2
if ordersize<1 then
ordersize=1
endif
ENDIF
sellshort ordersize shares AT close[1]-2 stop
endif
endif
endif
SET STOP ptrailing 5
Hi, I made the small changes in programming. Now, instead of launching the purchase or sale just after finishing the first 5 minutes, what it does is to launch a purchase order 2 points above the closing if it is positive, a sales order of the points below the closing Yes Is negative.
// Definición de los parámetros del código
DEFPARAM CumulateOrders = false // Acumulación de posiciones desactivada
DEFPARAM FlatAfter =173000
once ordersize=1
HoraEntradaLimite = 090600
HoraInicio = 090500
if Ordersize>15 then
Ordersize=15
endif
if Time >= HoraInicio and time <= HoraEntradaLimite then
c1 = open < close-2
if not onmarket then
IF c1 THEN
IF PositionPerf(1) < 0 THEN
OrderSize = OrderSize/2//+1
if ordersize<1 then
ordersize=1
ENDIF
ELSIF PositionPerf(1) > 0 THEN
OrderSize =OrderSize+2
if ordersize<1 then
ordersize=1
ENDIF
endif
buy ordersize*6 contract AT close[1]+2 stop
endif
c2= open > close-1
IF c2 THEN
iF PositionPerf(1) < 0 THEN
OrderSize = OrderSize/2//+1
if ordersize<1 then
ordersize=1
ENDIF
ELSIF PositionPerf(1) > 0 THEN
OrderSize =OrderSize+2
if ordersize<1 then
ordersize=1
endif
ENDIF
sellshort ordersize*6 contract AT close[1]+2 stop
endif
endif
endif
SET STOP ptrailing 5
The code is no different to Raul's above. Here it is again (with a few bits translated into English by Google). I thought it would be useful to attach the screenshot of the equity curve or the Backtest Report for everyone however I didn't know it wasn't allowed in the comments.
// Definition of parameters
DEFPARAM CumulateOrders = false // No accumulation of orders
// The position is closed at 4:30 p.m., local market time (London).
DEFPARAM FlatAfter =162900
once ordersize=1
// No new positions after the candle closes at 08:06
EndTime = 080600
// Market analysis starts in the candle 5 minutes closes at 08:05
StartTime = 080500
// Max orders
if Ordersize>11 then
Ordersize=11
endif
// Risk & multiplier contracts
n=1
// Conditions for analysis
if Time >= StartTime and time <= EndTime then
// Conditions to enter long positions
c1 = open < closeO-1
IF c1 THEN
IF PositionPerf(1) < 0 THEN
OrderSize = OrderSize/2//+1
if ordersize<1 then
ordersize=1
ENDIF
ELSIF PositionPerf(1) > 0 THEN
OrderSize =OrderSize+2
endif
// If the first day of 5 bar min is positive, we buy
buy ordersize*n shares at market
endif
// Conditions to enter short positions
c2= open > close-1
IF c2 THEN
iF PositionPerf(1) < 0 THEN
OrderSize = OrderSize/2//+1
if ordersize<1 then
ordersize=1
ENDIF
ELSIF PositionPerf(1) > 0 THEN
OrderSize =OrderSize+2
ENDIF
// If the first day of 5 bar min is positive, we sell.
sellshort ordersize*n shares at market
endif
endif
SET STOP ploss 3 //5
SET TARGET pPROFIT 12 //10
You can see clearly from the equity curve what you mean by whole months of nothing going on...just ticking along with small losses, then on a number occasions there is a run of 6+ winners in a row and then there is a nice spike up in profits. Almost like steps is you see what I mean. I would say approximately 20 of these "steps", followed by a "quiet period" of small losses, then another "step"
I've backtestest from 2 Jan 2013 until 20 Dec 2016 (just for making forward testing nice and neat 4 1-year periods). It's almost 200,000 bars. I can't attach files when I take a screen shot and I have even downloaded SnapMyScreen.
I've made a few changes to your first code (where the order size is reduced by /2). I've limited ordersize to 11 to be conservative. Also focused on minimising max drawdown.
n=1, Ordersize = 11, Stop = 3, Target = 12 (I know, the stop may be too close and I could miss profitable moves). I'll see how it goes and may go back to a stop of 5. Spread 1. I'm also including an Order Fee of EUR1 / contract.
Results seem promising: Gain 787.09%, % winning trades 26.54%, Gain/Loss 1.75, Max Drawdown 3,886. As to be expected a brilliant performance in Q4 2016 with nearly EUR30k profits (all building up to this). So far this year just seems to be "ticking over" with small losses.
I'll attach files when I can. It always says "HTTP error".
Hope this helps.
Hi jon. Yes, but look at the backtest as there are whole months going flat as of January and February 2016. Or in September, there isn´t strategy (at least that I know) that of a perfect money evolution, but at least in this when you lose, you lose little and when you win, the profits soar. I'm looking for someone with 200,000 bars to see how it behaves earlier and see if it's a solid strategy.
Thanks Raul. I get it mathematically now. Just slow getting my head around it after too much screen time. A big thing I wasn't factoring in (even though it was in plain sight) is the difference in the target (10) and stop (5).
As I said, I wasn't trying to be pedantic or difficult.
Thanks again. Great strategy.
Hola y gracias por compartir! Descubrí que en 5 minutos teníamos algo muy bueno! Pero no en el buen sentido del mercado ... ¿es posible devolver las compras / ventas para hacer pedidos positivos? Gracias de antemano