[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [amibroker] Robustness testing & Over-optimization checklist



PureBytes Links

Trading Reference Links




Hello Dave,
 
For #10, Kaufman is refering to the notion that the 
majority (68%) of the range of parameters should be profitable.  In a map 
of parameter values & associated returns, you want to see a plateau instead 
of just spikes.  Check out the 3D optimization spreadsheet in the Files 
section.
 
Also, the order of parameters is crucial because 
you should test the most importanat variables first, i.e. the ones that cause 
the largest changes in your results.  Focus on those first and then on the 
less important ones
 
Yes, the book is excellent, not just for the 
robustness section.  I haven't read his other trading system book yet 
though so I don't know if there are similarities.
 
HB
<BLOCKQUOTE 
>
  ----- Original Message ----- 
  <DIV 
  >From: 
  Dave Merrill 
  
  To: <A title=amibroker@xxxxxxxxxxxxxxx 
  href="">amibroker@xxxxxxxxxxxxxxx 
  Sent: Saturday, September 27, 2003 8:53 
  AM
  Subject: RE: [amibroker] Robustness 
  testing & Over-optimization checklist
  
  <SPAN 
  class=213014312-27092003>thanks, I think I made that original request, maybe 
  others too.
  <SPAN 
  class=213014312-27092003> 
  <SPAN 
  class=213014312-27092003>what do you make of these:
  <SPAN 
  class=213014312-27092003>g    
  gggggggggggggggggggggg
  <FONT face="Courier New" 
  color=#0000ff size=2>> 8. Will you test a full range of parameters.> 
  9. In what order will the parameters be tested?
  <SPAN 
  class=213014312-27092003>> 10. Are the parameters distributed 
  properly?
   
  <SPAN 
  class=213014312-27092003>why would you expect profitability for all parameter 
  settings? does the existence of some unprofitable settings invalidate a 
  strategy? <SPAN 
  class=213014312-27092003>what difference does the order in which parameters 
  are tested make? what does #10 mean?
  <SPAN 
  class=213014312-27092003> 
  <SPAN 
  class=213014312-27092003>did you like the book overall? would you say it 
  answers the robustness question well, beyond this 
  checklist?
  <SPAN 
  class=213014312-27092003> 
  <SPAN 
  class=213014312-27092003>thanks,
  <SPAN 
  class=213014312-27092003> 
  <SPAN 
  class=213014312-27092003>dave
  <BLOCKQUOTE 
  >
    I recall someone asking about 
    over-optimization and robustness testing.  Here's an excerpt 
    from...<FONT face="Courier New" 
    color=#0000ff> 
    <SPAN 
    class=213014312-27092003> 
    Book: "Smarter Trading: Improving Performance 
    in Changing Markets"
    
    Chapter: Making a Strategy 
    RobustAuthor: Perry Kaufman
     
    CHECKLIST FOR ROBUST TESTING
    PART 1: Deciding What to Test<FONT 
    face=Arial size=2>
    1. Is the strategy logical?2. Can you program all the 
    rules?3. Does the strategy make sense only under certain 
    conditions?4. Take a guess as to the expected results.
    PART 2: Deciding How to Test
    5. Choose the testing tools and method.6. Do you have enough of the 
    "right" data?7. Have you Included realistic transaction costs?8. 
    Will you test a full range of parameters.9. In what order will the 
    parameters be tested?
    10. Are the parameters distributed properly?11. Have you defined 
    the evaluation criteria?12. How will the output be presented?
    PART 3: Evaluating the Results
    13. Are the calculations correct?14. Were there enough trades to be 
    "significant?"15. Does the trading system produce profits for most 
    combinations of parameters?16. Did logic changes Improve overall test 
    performance?17. How did it perform on out-of-sample data?
    PART 4: Choosing the Specific Parameters to Trade
    18. Did the last test include the most recent data?19. Did you 
    choose from an area of broad success?20. Are profits distributed 
    relatively evenly over the tested history?21. Are the profits per trade 
    large enough to absorb errors?22. Did the historic results show any 
    large losses due to price shocks?23. Have you risk-adjusted the returns 
    to your acceptable
    PART 5: Trading and Monitoring Performance
    24. Are you following the same rules that were tested?25. Are you 
    trading the same data that was tested?26. Are you monitoring the 
    difference between the system's actual entries & exits?
     Send BUG REPORTS to 
    bugs@xxxxxxxxxxxxxSend SUGGESTIONS to 
    suggest@xxxxxxxxxxxxx-----------------------------------------Post 
    AmiQuote-related messages ONLY to: amiquote@xxxxxxxxxxxxxxx (Web page: 
    <A 
    href="">http://groups.yahoo.com/group/amiquote/messages/)--------------------------------------------Check 
    group FAQ at: <A 
    href="">http://groups.yahoo.com/group/amibroker/files/groupfaq.html 
    Your use of Yahoo! Groups is subject to the <A 
    href="">Yahoo! Terms of Service. 
    Send 
  BUG REPORTS to bugs@xxxxxxxxxxxxxSend SUGGESTIONS to 
  suggest@xxxxxxxxxxxxx-----------------------------------------Post 
  AmiQuote-related messages ONLY to: amiquote@xxxxxxxxxxxxxxx (Web page: <A 
  href="">http://groups.yahoo.com/group/amiquote/messages/)--------------------------------------------Check 
  group FAQ at: <A 
  href="">http://groups.yahoo.com/group/amibroker/files/groupfaq.html 
  Your use of Yahoo! Groups is subject to the <A 
  href="">Yahoo! Terms of Service. 







Yahoo! Groups Sponsor


  ADVERTISEMENT 









Send BUG REPORTS to bugs@xxxxxxxxxxxxx
Send SUGGESTIONS to suggest@xxxxxxxxxxxxx
-----------------------------------------
Post AmiQuote-related messages ONLY to: amiquote@xxxxxxxxxxxxxxx 
(Web page: http://groups.yahoo.com/group/amiquote/messages/)
--------------------------------------------
Check group FAQ at: http://groups.yahoo.com/group/amibroker/files/groupfaq.html



Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.