Edgeware FastBreak Pro Version 6.2 Uživatelský manuál Strana 1

FastBreakProVersion 6.xwww.edge-ware.com

10 Buy filter using a short/long Exponential Moving Average (EMA) crossover. This issimilar to the current EMA Buy filter except that the fund NAV is

11Note: You can load trading systems developed with Version 3 of FastBreak into thisnew version, however, you must input the beta /correlation index a

124.0 How FastBreak Pro is Different from Standard FastBreakOur intent when we designed FastBreak Pro was to automate development of mechanicaltrading

13to try, and this is why FastBreak Pro uses a very sophisticated genetic algorithm (GA) toevolve the parameter choices.Determining optimized paramete

145.0 InstallationNew users should first read the Standard FastBreak manual supplied with your purchase.That manual describes all the parameters and o

156.0 Technical Support and UpgradesNote: Do not call Investors FastTrack for technical support.Do not use the 800 order line number for technical sup

16 Verify that the FastTrack account number on this computer matches the accountnumber on the FastBreak CD Install FastBreak using the installation

177.0 FastBreak Pro OperationWe recommend that you read this manual in the order it was written; however, if youhave experience with Standard FastBrea

18To switch back to the “Standard” FastBreak screens described in the standard manual usethe icon on the far right of the icon tool bar:The following

192ndNote: the Optimize Family option is not available when using trendline Stops orBuy filters. See Appendix A for details.To prevent over-optimizati

2Disclaimer and License AgreementThe information that FastBreak Pro provides is only part of the information needed for agood investment program. Cons

20predictive ability. Normally, the OS date range immediately follows the IS optimizationdate range. FastBreak Pro does not require this typical order

21such as GAM and MAM, you will see text box ranges on the optimizer screen thatallow you to insert ranges for this third variable.You can change the

22often the best solution historically. This can be compared to the “buy the dips” techniqueof the late 1990s. The unfortunate problem is that, in a b

23If you check the Correlation box, FastBreak Pro will try a maximum correlation, betweenfunds to hold, for some strategies:The optimizer will try to

24will not work with all the ranking methods. See the Standard manual for a fulldescription.You can force FastBreak Pro not to use the option, to use

25The optimizer will determine the Top% value for your strategy using the range of valuesin the following boxes:In this example, a value between 10% a

26would result in only one day to calculate the curve fit. The curve fit equations requireup to three days to make the calculation, and FastBreak Pro

27Stop LossClick this icon to bring up the stop parameters page:Stop Begin DayThis option is used in the optimization process to suspend all Stops for

28a minimum number of calendar days is reached. The option is at the top of the screenwhere Stop optimization ranges are set.There is a single value i

29You are not required to use the second period. If you want a single stop value to be foundfor the entire time period a fund is held, check only the

3Table of Contents1.0 Preface...42.0 Introduction...

30Buy Filters ScreenClick on this icon to bring up the Buy Filters screen:Buy filters are checks that can be applied to a fund prior to purchase. See

31ROR is to be calculated. For example, FastBreak may determine that the fund to bepurchased should be increasing at the rate of at least 8% per year

32wants to force this filter to be used the Force Use box can be checked. This will makeevery strategy that the optimizer tries use this option. If th

33Genetic Algorithm ScreenNote: To better understand the following discussion you may want to go to Appendix Aand read the discussion on Genetic Algor

34Only one of these three parameters can be selected for optimization using the radiobuttons. If a parameter is not selected for optimization, FastBre

35some genetic algorithm books, but we have not found these large population sizesto result in improved system performance using the 50% Survivor Sele

36described in the Appendix. Note: Activation of the Maximize Robustness option willtriple run time; however, we believe it is critical to building go

37database can be used. Note: If using the Minimize Beta option, the Beta Buy Filtershould also be used on the Buy Filter screen. For example, if a us

38The information in this file will be covered later when the optimizer is covered.Using this option can result in quite a large file. Use with care.T

398.0 Running FastBreak Pro OptimizationThe GA optimizer works interactively with a special version of Standard FastBreak.Standard FastBreak provides

41.0 PrefaceNote: A PDF copy of this manual is on your installation disk. We recommend you puta copy on your computer so you will always have a copy a

40Note: Your screen may look different depending on your monitor screen resolution. Ifyou have a very low resolution screen, the far right columns may

41Buy FiltersO ROR Buy FilterE EMA Buy FilterP Parabolic Buy FilterR RSI Buy FilterC Correlation Buy FilterB Beta Buy FilterX EMA Crossover Buy Filter

42Note: It may take several seconds or a few minutes to pause execution. If you have therobustness option checked, FastBreak may run up to 3 additiona

43Notice that there are two lines on each graph. One line is for the #1 strategy identified byFastBreak Pro (based on Adjusted performance in the Best

44The Out-of-Sample (OS) Results screen adds a new line at the end of each generation.When we refer to “Average” values, we are referring to a simple

45ANN SD The standard deviation of the Out-of-Sample annual returns for the top tensystems. The larger the value, the larger the variation in Out-of-S

46Saving StrategiesThe parameters for the top ten best systems can be saved after a run has finished, or theprogram can be paused during optimization

47You can then see the specific strategy parameters found and you can execute FastBreakPro in the standard manner to view Detail results or FNU files.

489.0 ExamplesWe have included example files on the installation CD. You can use the files directlyfrom the CD, but we recommend copying the files to

49hold three funds because we set the minimum and maximum range for # funds held to 3.We are optimizing on Annual return, but have specified an MDD go

52.0 IntroductionEdge Ware, Inc. released the initial Standard FastBreak version in early 1996 to allowinvestors to design, evaluate and trade fund ro

50future market conditions. The reason we reserve some recent out-of-sample market datais to test the optimized parameters. We are looking for the poi

51The OS MDD is a little disappointing because we optimized with a 15% MDD, and wesee that the OS MDD is typically in the 17-22% range. There are some

52After you choose the generation number you will be asked to provide a name. We willuse the name Exp and FastBreak will provide the extensions (Exp0.

53In this example, we chose system # 10 in generation 9 because it was the best in the Out-of-Sample screen. Should we always automatically choose the

54and evaluate them separately, but we will use the FastBreak Pro family optimizationoption to determine the best combination of funds. We build a tra

55This trading strategy has very good performance in the OS period. The results are evenmore impressive compared to the major market indexes during th

56We see that all 10 new families (ET0 through ET9) are now available in the Familiesselection table. At this point we execute the strategy just like

57Example 3 – Trading StocksOne of the most exciting and aggressive uses of FastBreak Pro is developing stocktrading systems. Note: Trading individual

58an inaccurate indication of performance when performance results are measured overa very short trading period. For this reason, it is important to h

59Create the DFT files for generation 14 using the method in the first two examples. Now,load strategy 5 DFT into FastBreak, change the strategy start

63.0 Upgrade NotesMaintaining Existing FastBreak Trading StrategiesIf you have existing FastBreak strategies that you like we recommend that you keep

60Example 4 – Building Market Timing Signal FilesRead the chapter on building market timing signals in the Standard FastBreak manual.That chapter will

61Each day you can load this DFT file and execute it just as you would any trading system.If there is a market buy or sell, the signal file will be up

6210.0 Suggestions for Building Better SystemsThe optimization parameter defaults in FastBreak Pro will give you good results. In thischapter we show

63Keeping Track of Optimization RunsIt is useful to keep a simple record of optimization runs. Although it is a simple matter toreload the restart fil

64return (note the false sharp peak at generation 3). However, we are seeing diminishingimprovement in the IS results, and this is a reasonable place

65hold more funds the variations in OS performance among the ten systems is reduced. Thesingle fund system has the system with the best OS result but

66"Stale"monthsOSReturn,%/year0 37.76 40.718 41.8These results are very encouraging because all systems are comparable and the “older”system

67What is very clear is that in the early to mid 90’s, it appears that your chance for out-performance was much better. The ratio hit a low in the Jun

68Trading International FundsWe built a family of 34 international funds that can be traded with either no, or lowtransaction fees. We examined the im

69discover ways of choosing from among the “best” trading systems provided by FastBreakPro at the end of each generation.When FastBreak Pro completes

7trendline that a human brain can draw in a few seconds is a daunting task; however, thereis a tendency when manually drawing trendlines to “force” th

70It can be argued which of the ten systems would have been chosen to be traded. The out-of-sample MDD was very similar in all cases. In this particul

71System OS Return FNU OS Post OS FNU PostNumber Annual, % Annual,% Annual, % Annual, %1 38.5 44.6 17.7 202 28.7 37.7 23.3 21.13 5.8 33 32.2 32.14 41.

72Generation 8 was selected for detailed analysis. Here are the results (using FNU equitycurves):System OS Return Post OSNumber Annual, % Annual, %1 2

73Reversing the In-Sample and Out-of-Sample Data PeriodsFastBreak Pro allows the OS data period to be earlier than the IS data period. The moretraditi

74history. We found that optimization on this recent bear market data improved riskperformance.The Effect of MDD Objectives on PerformanceAll investor

75happened when we tried to build a system with a 10% MDD. We have put an option inFastBreak Pro that applies a penalty to systems that activate sever

7611.0 Frequently Asked Questions and Common ProblemsQ) I crashed my hard drive (got a new computer, new laptop etc.) and I need to reinstallFastBreak

77a temporary family file to use. This will create confusion if multiple executions aretrying to read and write to this family file.Q) I created a new

78Q) Should I ever “over-ride” the trading system?A) Usually the answer is no. One very good reason to stop trading a system is if thesystem starts to

79Appendix A -- Technical DiscussionWhat are Genetic Algorithms?Genetic Algorithms are a mathematical method used to solve hard optimization problems.

8Changes between Version 4 and Version 5Version 5 of FastBreak Pro dramatically improved the optimization time when usingFNU data files. FNU files are

80Adjusted Performance = 20% x 15/18 x 10/14 = 11.9%This calculation is made for all systems (chromosomes) in a generation. All the adjustedperformanc

81In this example, the parameter values are all increased by 5%. The strategy performancefrom this run is saved. Then a second evaluation is made, but

82Since 0.86 is greater than 0.85, no robustness adjustment is made to the original 24%performance.Which method, Average or Lowest, is better? Our res

83usually results in a reduction to the variable being optimized (constraints such as MDD,or switches per year have the same effect). We are willing t

84The time it takes to build the trendlines is dependent on the following: Number of family members Size range for trendlines to be built. If you us

9while large cap stocks are not. FastBreak allows the use to combine multipleindexes (or any stock, fund, FNU file) into a family. This allows the use