Grid orders with one combined stop loss and limit, can it be done?
-
Awesome Nicolas!
Since it a trend following system my guess is that the hit rate will not be very very high, so in that case we need big profits when winning and small losses when losing. And that is exactly what I think this system can do 🙂To start trailing when the floating profit reached at least a risk/reward x factor sound very good! ProRealTimes term “profit percent step”, is actually a new term for me because of my background from MetaTrader, so I’m not sure how ProRealTime calculate that.
But I think as long as there is enough space between market and trailing stop, I think thats perfect! Enough so the market can do its magic 🙂 but of course not to big so the market can do its other magic resulting in a loss (But that is of course taking care of with the “To start trailing when the floating profit reached at least a risk/reward x factor sound”). Its a balancing act.Maybe ProOrder will require a profit percent step because of the basket of orders?
Maybe the Trailing stop moves could be based on volatility, maybe with ATR? When high volatility, means bigger distance from market, and when low volatility – smaller distance from market?
Or maybe maybe distance based on time, first bigger distance between trailing stop and market, and later on smaller distance. It could make sense if a big move means high volatility and then the following consolidation means lower volatility, but Im not sure if it means that 🙂I like sometimes to use PSAR as a trailing step, I think it calculate volatility as well. But sometimes it don’t work, in my testings it has also lead to curve fitting I think.
My best suggestion is that as long there is enough distance between trailing step and market, anything will be very fine! 🙂
And maybe if the Trailing stop moves could be based on volatility, maybe with ATR?All the best!
I’m glad to see all these suggestions and ideas!
As for profit protection, standard deviation from the floating profit when R:R x is reached sounds like an amazing idea since the worst enemies of this systems are retracements and whipsaws, I would love to try it with a bollinger band cross.Regarding how I pick my entries Nicolas is right, I base my decisions on technical analysis mainly on the H4 timeframe, using MACD, Heiken Ashi, Heiken Ashi Smoothed, MA, a CCI histogram, a custom MTF ADX matrix, MTF ATR for reference of determining number of pips for SL and target. I used to trade a basket of currencies which you can read more about here, though now I have replaced the IXOAH with MACD and switched ADX indicator;That post is over 2 years old but I have not changed my trading that much since then, since I started using PRT just over two months ago I still check my MT4 charts along with PRT because it’s easier for me to spot the trades there until I have spent more hours getting accustomed to the PRT charts.While testing the system I started it on GBPCAD on Monday last week at a very good level which I posted here;The attached screenshot shows how my MT4 chart looked and I guess it’s pretty obvious why I took it when looking at the indicators (note that the ADX matrix and ATR average doesn’t show the data for that time but from Friday’s market close), the exit may have been premutare but since the MACD started showing signs of a reversal I called it quits.I think it would be rather challenging to automate my entry strategy in particular to avoid ranging markets when losses can rack up quickly. However creating a market scan to find the oppurtunities and then verify the setup manually and then starting the system can be very useful, alternatively semi-automating entries creating an alert when the setup occurs and requiring a manual confirmation to start the system.Unrelated to the above the system can be used when anticipating market events causing heavy volatility such as central bank rate decisions with uncertain outcome and market direction. In those scenarios we want to start the system one two charts of the symbol, one going long and the other going short. If price takes of in either direction with only moderate whipsaw and retrace we will lose 1 % on one side but strike gold on the other! Recent opportunities include the Bank of Japan rate decision on virtually any JPY pair but let’s take a look at GBPJPY which would have racked up 500+ pips within hours which would give us over 37 % profit or even more if holding for a day or two and don’t get me started on the 1200+ pips moev on the Nikkei at the same time! A similar opportunity occured on 3rd May after the RBA rate decision, smaller amount of pips but still nice profits 🙂We have many great things to come my friends!Hello guys,
I have spent part of the weekend trying to figure out which parameters to use for a market screener since finding good setups is a little different when using a coded system rather than just glancing at it and recognizing a setup. My strategy for picking entries is pretty good but there is probably a fair amount of room for improvement, that said some of the exiting screeners on forum might be more useful, I’m looking to test MikcGC’s Bullish/Bearish and the TDI screeners this week and gain some insight.
Nicolas I have continued the live testing and everything looks good with the exception of JPY pairs where I seem to get stopped out very soon after entry by very small retraces, can this be some sort of decimal issue or that the system count amount of JPY lost rather than the base currency? Please have a look…
defparam preloadbars = 0 //parameters accountbalance = 10000 //account balance in money at strategy start riskpercent = 1 //whole account risk in percent% gridstep = 20 //grid step in point amount = 1 //lot amount to open each trade //first trade whatever condition if NOT ONMARKET AND close>close[1] AND STRATEGYPROFIT=0 then BUY amount LOT AT MARKET endif // case BUY - add orders on the same trend if longonmarket and close-tradeprice(1)>=gridstep*pipsize then BUY amount LOT AT MARKET endif //money management liveaccountbalance = accountbalance+strategyprofit moneyrisk = (liveaccountbalance*(riskpercent/100)) onepointvaluebasket = pointvalue*countofposition //GRAPH mindistancetoclose as "d" //stoploss trigger if onmarket then mindistancetoclose =(moneyrisk/onepointvaluebasket)*pipsize SELL AT positionprice-mindistancetoclose STOP endif
- You must be logged in to reply to this topic.
Summary
This topic contains 307 replies,
has 1 voice, and was last updated by 
1 year, 10 months ago.
Topic Details
| Forum: | ProOrder: Automated Strategies & Backtesting |
| Language: | English |
| Started: | 04/14/2016 |
| Status: | Active |
| Attachments: | 106 files |
