| Category | Prediction Rate | % Correct | Notes |
| Direction | 10/12 | 83% | 12/12 disregarding minor closing range |
| Range | 7/12 | 58% | 11/12 disregarding minor closing range |
| Significant High/Low* | no instances | Threshold raised to 70% | |
| Unfilled Gap** | 1/2 | 50% | |
| Next Day Gap** | 2/2 | 100% | |
| Gap Fill+ | 1/1 | 100% |
** An unfilled gap occurs when the market gaps and never fills. The next day gap determines if the condition correctly predicted the gap.The unfilled gap and next day gap tests are measuring times the odds of a positive low or negative high were >= 14%.
+ When the odds of a negative high/pos low <= 3% and the market gaps in the low odds direction, the gap fills.
First to deal with yesterday's tests:
1) Since the odds of a negative high (unfilled gap down) are <=3%, if the market gaps down, the gap fills. Not applicable
2) Since the odds of a significant high >= 66%, the market makes a significant high >0.5% tomorrow. NO.
The failure of the 66% threshold to predict a significant move the next day leads me to believe that threshold has to be higher. I'm resetting that test and raising the threshold to 70%. On the other hand, I noticed recently that whenever that high of a figure appeared, big gains did happen in the market in the near future, so perhaps it becomes useful to predict that, but I haven't tracked it.
Direction was correct today but range was wrong, but considering that the Dow traded negative most of the day, I'd say it wasn't off by much, so I'll give it credit in the notes.
Monday looks similar to today in that the odds are higher of an up day but the range was generally negative. If we are to get some sort of rule out of this, I wouldn't count on a big move Monday, but the direction looks positive on a closing basis. That seems to be trumping the range prediction recently.
No comments:
Post a Comment