"I think he is still posting but only for PAL users. Regardless, I think you are talking about the indicator part because I have seen the posts. I have also seen posts with patterns. The last one about TLT was right on the post with perfect timing:" Actually I thought the indicator value was a result of pattern research ?!? How does it work if not like this ? I'm much interested in PAL and found the bog useful mixed with other analysis but gave up on the idea of ordering a trial after reading PAL requires much time to use and the 1 year license is for 1 computer only. As i work both from home and the office and run a business besides trading, it's not convenient at the moment, but the prospect of getting the lifelong license and get someone to help me on those manipulations is a distinct possibilty. Yet, i'm not convinced the patterns have any predictive use, as previously mentionned, the sample sizes are usually very low. I'm expecting Harris' "Short-term Trading with Price Patterns" any day now to have a closer look in the issue, but any input here will be welcome. Also if the system is good, do you understand why Michael Harris seems to be having difficulties to sell 50 lifelong licences ? Price has been reduced around 25% over the last few months, and some (most ?) licenses are still available Besides find below the test I was refering to : PAL DELTA OVER 10 SIGNIFICANCE OVER 8 date %profit ticker %Long %short delta significance open target SL result March 5 3% CAT 57,18 40,06 17,13 18 37.49 333 -1 March 5 3% HPQ 58,1 41,2 16,9 14 25,18 25,93 24,42 -1 March 5 3% MCD 57,01 42,53 14,48 11 99,42 102,4 96,43 -1 March 5 3% VZ 40 59,5 -19,5 10 333 -1 March 5 3% JPM 41,8 57,04 -15,24 21 40,71 39,48 41,93 1 2% TLT 60,98 39,02 21,95 10 116,6 118,93 114,26 -1 March 2 2% XLF 40,72 57,22 -16,49 9 14,92 14,62 15,22 1 feb 24 3% WMT 59,64 40,06 19,58 16 58,75 61,72 56,99 open feb 24 3% JNJ 56,27 43,39 12,88 14 64,76 66,05 62,81 open feb 24 3% CAT 55,43 44,35 11,09 22 116,29 119,78 112,8 -1 feb 24 3% AXP 54,76 41,43 13,33 10 52,93 54,52 51,33 1 feb 21 3% KFT 61,02 37,29 23,73 8 38,64 39,8 37,48 open feb 21 3% KO 56,63 43,37 13,26 13 68,88 70,94 66,81 1 feb 21 3% TRV 45,07 54,74 9,67 27 59,95 61,74 58,15 -1
luisHK - can you post the link to the results. Hard to understand what the numbers are all about without the link.
The original signals come fromthe PAL blog: http://www.priceactionlab.com/Blog/...probabilities-for-dow-30-stocks-march-5-2012/ than again on March 2nd, feb 24th and Feb 21. I was planning to run it for 1 year on signals with delta over 10 and significance 8 and higher but Harris stopped posting the signals. So far it's 7 losses and 4 wins. Besides I have the info much clearer on excell but this website doesn't allow excell attachment.
Thanks. This indicator is like an oscillator so you do not know the values after that day. I don't think he posts that as a service but as an example of his program output. Did you read the commnets after the results? "Comments: There are a few different ways of interpreting the p-indicator results. The best way of interpreting the results is in conjunction with the results of other indicators. No indicator alone can consistently track market movements. Please note that the p-indicator calculations change on a daily basis and as a result the signals derived from it. " The signals change on a daily basis like those of any other indicators, a MACD or RSI. I would not trade the signals of any single indicator or pattern. Edit: I think delta below 20 is very risky. A value of 10 is like trading long when RSI(14) = 45. You want at least delta = 40 in this case or RSI(14) = 30. This will also depend on the trend bias of each stock. If the trend bias is high you want larger delta.
Sure I read the comments, but if the indicator is supposed to be of any value, (it seems to me at least ) it must have a predictive effect on the date PAL calculates the stats and the indicator. Hence I'd be curious of its reliability on a 1 year timespan. Besides thanks for your input on how using more effectively the indicator.
No problem. I think I agree about the predictive value but that should hold given that the indicator stays at or above the vaue it had at the time of the prediction.