Statistiche › Pattern tradabili › Technical rules
The best technical rule over a century was significant, and nothing has been since 1986
| Sample and universe | Sharpe | Nominal p | Adjusted p | Survives |
|---|---|---|---|---|
| Dow 1897 to 1996, full universe | 0.82 | 0.000 | 0.000 | Sì |
| Dow 1897 to 1986, full universe | 0.91 | 0.000 | 0.000 | Sì |
| Dow 1897 to 1996, moving averages only | 0.39 | 0.000 | 0.000 | Sì |
| Dow 1987 to 1996, full universe | 0.87 | 0.000 | 0.903 | No |
| Dow 1987 to 1996, moving averages only | 0.28 | 0.127 | 0.721 | No |
| S&P 500 futures 1984 to 1996, full universe | 0.66 | 0.000 | 0.987 | No |
| S&P 500 futures 1984 to 1996, moving averages | 0.23 | 0.165 | 0.702 | No |
Best rule Sharpe ratios from a universe of 7,846 technical trading rules, with nominal p values and p values adjusted for the fact that 7,846 rules were searched. Buy and hold Sharpe over 1897 to 1996 is 0.034. Every sample that ends in 1986 passes the adjustment. No sample after it does.
Testing 7,846 rules on a hundred years of Dow data, the best earns 17.2% a year and survives a data snooping correction. On 1987 to 1996 the same universe produces nothing significant, and on US futures only two of seventeen contracts survive out of sample.
Solo per membri
Il resto di questa pagina è per i membri
Below this point there are 6 sections and 3 named sources, roughly 800 words of it. Every figure carries the source it came from and the date the data is from.
Puoi continuare a consultare gratuitamente ogni statistica della libreria. L'introduzione e il riepilogo sono sempre accessibili.