Sorry, I can't comment on that as I have a very narrow focus on my trading, one Indice only, but you may be correct. Backtesting even just one indice is time consuming as well on SC, however as with all things it has its pros and cons. Trouble with CFD's is their pooled accounts (unsafe IMO) and their high fees.
You can do that too. Save analysis project with filter to explore only watchlist 0. Every symbol that will have new tick it will be moved to watchlist 0 Event == true will trigger exploration on watchlist 0 Code: RequestTimedRefresh( 1 ); /* you have to detect when exploration is complete using addtocomposite() to count the symbols analyzed and match that with numUpdated */ AnalyzedSymbols = Nz( Foreign ( "~~~AnalyzedSymbols" , "X" ) ); numUpdated = Nz ( StaticVarGet ( "numUpdated" ) ); analysisComplete = AnalyzedSymbols == numUpdated; if ( analysisComplete ) { event = 0 ; UpdatedSymbols = ""; numUpdated = 0; SymbolList = CategoryGetSymbols ( categoryMarket, 0 ); for ( n = 0 ; ( symbol = StrExtract ( SymbolList, n ) ) != "" ; n ++ ) { CategoryRemoveSymbol( symbol , categoryWatchlist, 0 ); LastRecorded = Nz ( StaticVarGet ( "LastTrade" ) ) ; Last = GetRTDataForeign( symbol , "Last" ); if ( Last != LastRecorded ) { event = 1; CategoryAddSymbol ( Symbol, categoryWatchlist, 0 ); StaticVarSet ( "LastTrade", last ); UpdatedSymbols += symbol + "\n"; numUpdated++; } } StaticVarSet ( "numUpdated", numUpdated ); if ( Event ) { ShellExecute( "C:\\Program Files\\AmiBroker\\Scripts\\RunAnalysis.js", "", "" ); StaticVarSet ( "Executed", 1 ); } } Title = "AnalyzedSymbols = " + AnalyzedSymbols + "\nNum Updated = " +numUpdated ;
Shall I assume the "for life" equates to as long as you keep the balabce at $5,000 or above? Not knocking the deal for if you like the platform it is a great deal.
I've used it for over 10 years but don't listen to me or anybody else as the only way to really evaluate any program is to use it. So download a trial version (30 days), as well as the Users Guide, and kick the tires. In addition, you can get your questins answered from AB Support and on the Yahoo forum by experienced users and the developer. I think you will find that with respect to system/indicator development/testing, speed, ease of programming, etc. that it is unmatched. Where I would rate it mediocre is in the area of charting and built-in charting tools which depending on your trading style you might want to complement with other programs, including broker trading platforms.
I think it is excellent value for the money. Easy portfolio back-testing, WFO ( BTW I do not use it because I do not believe in it), excellent graphics with virtually no bugs, a little difficult language to learn for beginners but for me it is more than I can ask for. I plan to use it with simple AFL generated by price action lab to develop position trading systems for ETFs along the lines done by someone I know who works for a hedge fund.
Thank you very much, apfx! Never thought about that approach. Will try it out next week with real-time data. Although from looking at the code I get the idea how it will work out. Looks promising. Good input. Thanks again!
Hm, IMO the opposite is the case. Charting is very flexible and less stiff like that of TS/MC. Also for example like the chart layering via Plot function's z-order shift (like layers of an onion). And where else can you create chart animations like these ones http://www.youtube.com/watch?v=1U93Vq7n1Sc http://www.youtube.com/watch?v=CCilJYA2HsA Built-in charting tools? Not sure what is missing as I don't use stuff like Gann, Fibos (which are built-in) or what else there is. Are you talking about patterns? For example a while ago I saw this video about harmonic pattern (personally I'm not into that stuff) http://www.youtube.com/watch?v=IHGlrBwV55E So if something is missing it can be created very easily by coding it.
IMO, relative to other programs charting is a bit clunky starting with the cursor and extending to an inability to cleanly display and properly handle P&F. Charting tools are on the weak side from simple pitchfork without warning lines/internal divisions to weak execution of the other areas that you do not use (i.e., Gann, Fibonacci, harmonic patterns, etc.). In short, the program is strong on system development and testing and weak in the charting area that is used to a greater extent by discretionary traders. I use a lot of charting rather than systems which is why I find it essential to backstop the program with other software. Certainly in principal some (not all) of the charting tools can be created in AFL but for most this would be a very time intensive task even if the person has the required programming skills and, imo, is not worth the effort because other alternatives exist, including some very good free resources.
Hm, I'm a discretionary trader also. And mainly I've moved to AB because of the flexible charting as you can create everything by your own and because of the exploration environment. Have you made suggestions to their feedback center? Again I don't use stuff like P&F, pitchfork, gann and what not. What is the exact difference to others? Could you please provide a screen capture comparison? Maybe then I or others could better follow you.
From what you've said I have no doubt that AB's basic charting is adequate for your purposes both in terms of charting tools and AFL code. For more sophisticated charting one can, as you suggested, program the charting tools or use other progarms as backstops. Certainly some things such as price-based charting which requires control of the x-axis is not possible with the available AFL. In any case, not all have the programming skills, interest or time to do what is usually a non-trivial task of creating a new or enhanced charting tool. I don't know what you mean by "difference to others". If you want to pursue this contact me offline.