Monday, December 5, 2011

Targets and Things

Kevin, I will answer your questions in this seperate post. 
This board has a fairly severe length limitation on comments.

I'm using the cycle chart right now for examining randomness. I've got it cluttered with a lot of extraneous lines. I will clean it back up in a few days and re-post it if you like. Why don't you just keep your own chart? Why don't you ever post any of your charts? Or some of your backtest spreadsheets? I think people would find this interesting.

I see the target you are talking about. TARGET (12/21, 61.51). Those type targets are not truly Hurst targets. They come from a guy named Stevenson who wrote a book in 2004 "Precision Trading With Stevenson Price And Time Targets" Supposedly they are based on Hurst's work, but they seem like an over-simplification.  These Stevenson targets tend to be more reliable as to time and not so much as to price.  I wouldn't bet the farm on them. Though I know you never bet the farm on anything.

This is a very good idea you have of assigning a numerical description to the position in the upleg or downleg. Keep it very simple though. Why not just +1 for the first day of the upleg, +2 for the second day, and so on.... -1 for the first day of the downleg, -2 for the second day, and so on...

Using this numbering system you could then interface to the intraday chart with rules.  For instance today would have been +7.  This is getting fairly high up in the upleg.  A rule might be at plus 7 you begin looking to short any hod signal to hold intraday only.  Since your "Either/Or" signaled the HOD this morning, you would have entered short there and closed the trade sometime before the close.  As the upleg ages you could feel safe, in holding the trades longer, overnight or even longer, maybe at +10 or +11 and higher.  This is very rough, just to give you an idea.  I'm sure you can refine this, and will want to add more conditions than just the upleg count.

To estimate the 8-day and 16-day cycle bottoms is very easy, just a matter of counting bars.  But you have to make some estimate (guess) of the cycle length.  Right now the dominant cycle is oscillating at average 38.6 days over the last 12 months.  So we use this average.  This would make the 8-day bottom due to appear Thursday or Friday.  The 16 day bottom would be due almost the same time, maybe a day or two later. This could be a very minor reaction, a touch of the middle line at worst.  From there we should go to the top of the cycle, which should hit around 12/21.  If the 8-day 16-day bottom showed up late, we would expect the cycle top to show up late by the same number of days.

Good luck. Happy birthday if I don't communicate with you again before then.