Q No precision, good or bad, can be ascribed to a test method unless instrucment and observer as a combination show statistical control. This is so regardless of the cost of testing equipment. --- Out of the Crisis (W. Edwards Deming), P. 269 UQ
Q A stable process, one with no indication of a special cause of variation, is said to be, following Shewhart, In Statistical Control, or Stable. It is a random process. Its behavior in the near future is predictable. --- --- Out of the Crisis (W. Edwards Deming), P. 321 UQ
Q Courses in statistics often commence with study of distributions and comparison of distributions. Students are not warned in classes nor in the books that for analytic purposes (such as to improve a process), distributions and calculations of mean, mode, standard deviation, chi-square, t-test, etc. serve no useful purpose for improvement of a process unless the data were produced in a state of staistical control. --- Out of the Crisis (W. Edwards Deming), P. 312 UQ
One more note. Steve has stated several times to review acrarys work. Good advice again. And Acrary backtests. Food for thought. peace axeman
steve i am using excel for single market patterns. i tried several times to get the correlation search going, but it never paid off. instead i have build a little backtesting sheet and immediately test new features. i found this to be quicker than correlation analysis, most probably because i never made it get me somewhere. t-test serves as a co-check for avoiding overfit. in correlation analysis i found that many things beat random, but nothig significantly (i saw one exception to this - but i found that using the direct backtesting approach) excel's biggst advantage is that it takes so little time to implement things. i find it great. though we do all things finally in c++ and have a look at tradestation, i like the easyToGo aspect of excel. how many new things do you typically find per quarter? peace
an example. coffee. holding period: 1 day. 2 parameters. 199 trades. ttest: 0.00. hitratio: 61%. payOff: 1.61. modified sharpe ratio: 0.88. return p.a.: 16%. vola ann.: 18%. MDD: 17%. 1987 1988 -6% 1989 15% 1990 9% 1991 8% 1992 6% 1993 17% 1994 10% 1995 50% 1996 -3% 1997 25% 1998 32% 1999 13% 2000 12% 2001 28% 2002 22% 2003 19% 2004 00% i have it on papertrading to see how it acts in real world. peace
http://www.trade-ideas.com/History.html?O=3_3d_0&WN=FREE+High/Low+Ticker If you can't see the History from this free alert window, go to TradingScience where a two-week trial is offered. http://www.tradingscience.com/3rdpartyresources.html
Not following you here Acrary has posted backtesting results on numerous posts. I also test against random as well..... but backtesting is my core tool. peace axeman