Morning inversion of Ftsemib with RSI and previous day body size filter.

Francesco78 · 2017-03-21T14:20:30+00:00

Hello, I just started practicing with prorealtime automatic trading. It is great to have discovered this community. I wanted to show you a somehow interesting strategy based on taking the opposite direction of previous day in the morning between 09:00 and 09:30 on Ftse mib. I then add some filter in order to avoid taking the position in an overbought or oversold environment unless there is a momentum defined by the positive or negative local derivative of RSI. Furthermore I added a filter based on the size of the body of the previous day candle. Will be great if you tell me what you think about it. Ciao // Definition of code parameters DEFPARAM CumulateOrders = False // Cumulating positions deactivated timeenter = time = 090000 timeexit = time = 093000 b = 175 size = 5 // Conditions to enter long positions c1 = (DClose(1) 40) c5 = (indicator1[1] > indicator1[2]) c6 = (indicator1[1] = DOpen(1)+b) IF c2 AND c8 AND timeenter THEN SELLSHORT size PERPOINT AT MARKET ENDIF if shortonmarket AND timeexit THEN EXITSHORT AT MARKET ENDIF // Stops and targets SET STOP pLOSS 300

Viewing 4 posts - 1 through 4 (of 4 total)

Author

Posts
03/21/2017 at 3:20 PM #29297 quote
Francesco78
Participant

Master
Hello, I just started practicing with prorealtime automatic trading. It is great to have discovered this community.

I wanted to show you a somehow interesting strategy based on taking the opposite direction of previous day in the morning between 09:00 and 09:30 on Ftse mib.

I then add some filter in order to avoid taking the position in an overbought or oversold environment unless there is a momentum defined by the positive or negative local derivative of RSI.

Furthermore I added a filter based on the size of the body of the previous day candle.

Will be great if you tell me what you think about it.

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

timeenter = time = 090000
timeexit = time = 093000
b = 175
size = 5
// Conditions to enter long positions
c1 = (DClose(1) <= DOpen(1)-b)

indicator1 = RSI[14](close)
c3 = (indicator1[1] < 60)
c4 = (indicator1[1] > 40)
c5 = (indicator1[1] > indicator1[2])
c6 = (indicator1[1] < indicator1[2])

c7 = c3 OR c5
c8 = c4 OR c6

IF c1 AND c7 AND timeenter THEN
BUY size PERPOINT AT MARKET
ENDIF

if longonmarket AND timeexit THEN
SELL AT MARKET
ENDIF

// Conditions to enter short positions
c2 = (DClose(1) >= DOpen(1)+b)

IF c2 AND c8 AND timeenter THEN
SELLSHORT size PERPOINT AT MARKET
ENDIF
if shortonmarket AND timeexit THEN
EXITSHORT AT MARKET
ENDIF

// Stops and targets
SET STOP pLOSS 300
```
proreal-149010167484cpl1.jpg
03/27/2017 at 2:03 PM #29982 quote

Nicolas
Keymaster

Legend

Hi Francesco, are you sure you included the spread in your own tests? I did 200k bars backtest with 1.5 spread on MIB40 1€/point and I do not get the same result as you?

MIB40-trading-strategy-backtest.png

04/09/2017 at 9:28 PM #31544 quote

Francesco78
Participant

Master

Hi Nicolas, I just saw your answer, sorry, yes I included the spread, I re run it and I got results similar to my first test. What I find weird is that on my platform, 100,000 half an hour candles go back to Apr 2015, while I see that yours 200,000 candles go back much further than double mine.

Anyway I have just posted a refined version of this code on the forum. Would be great if you could have a look to that as well.

04/09/2017 at 9:34 PM #31545 quote

Francesco78
Participant

Master

Pls also note that I did the backtest in tick by tick mode.

Regards

Francesco
Author

Posts