2 replies to this topic

[Woody]

Aire.......here is an excerpt from an email response from Tom McLellan, would be interested in your opinion re the possibility of a 20 wk cycle low late Nov......... There are a couple of points worth noting about these cycles. The first and most important is that the precise period length is not 20 nor 40 weeks, but actually slightly less that those. It is just easier to say "40-week" than 38.3-week. My own work shows it running at a period of about 185 trading days. Most other analysts also do not understand the phase-shift phenomenon in the way that we do, perhaps because they have not done the lengthy research I have done. It is my contention that the 40-week cycle underwent a phase shift in late 2005. Past phase shifts have occurred about every 6-8 years, and so this is roughly on schedule. Past phase shifts have also shifted by varying lengths of time, and this one just happened to shift by a period approximately equal to half the length of the 40-week cycle. For that reason, if someone does not know about the phase shift phenomenon (or does not believe in it), then it is reasonable to conclude that we are still on the old schedule. And indeed, that may turn out to be the right conclusion. For now, there is no conflict in saying that the next 20-week bottom is due in late November, since every other 20-week low is a 40-week low. Whether it is as significant as the August low, and in what ways it might be significant, are secondary points in my view.

[airedale88]

Aire.......here is an excerpt from an email response from Tom McLellan, would be interested in your opinion re the possibility of a 20 wk cycle low late Nov.........

There are a couple of points worth noting about these cycles. The first
and most important is that the precise period length is not 20 nor 40
weeks, but actually slightly less that those. It is just easier to say
"40-week" than 38.3-week. My own work shows it running at a period of
about 185 trading days.

Most other analysts also do not understand the phase-shift phenomenon in
the way that we do, perhaps because they have not done the lengthy
research I have done. It is my contention that the 40-week cycle
underwent a phase shift in late 2005. Past phase shifts have occurred
about every 6-8 years, and so this is roughly on schedule. Past phase
shifts have also shifted by varying lengths of time, and this one just
happened to shift by a period approximately equal to half the length of
the 40-week cycle. For that reason, if someone does not know about the
phase shift phenomenon (or does not believe in it), then it is
reasonable to conclude that we are still on the old schedule. And
indeed, that may turn out to be the right conclusion.

For now, there is no conflict in saying that the next 20-week bottom is
due in late November, since every other 20-week low is a 40-week low.
Whether it is as significant as the August low, and in what ways it
might be significant, are secondary points in my view.

woody, it would take a very long response to fully answer this question. i'm also having problems loading charts to image shack so i can't offer any charts that would help shorten the answer. the phase shift mentioned occurred in 2005. late april 05 was a nominal 40 wk low, and a visually prominent low in oct, approx 25 wks from the april low, is labeled by some analysts as a phase shifted 40 wk low. Hurst methods identify that oct low as a cycle straddle, a smaller cycle low with a sudden amplitude jump. to confound things, the next Hurst based non phase shifted 40 wk low from the late april 05 low was due in january 06, and on broad index charts there is no visually prominent low (an even more important nominal 80 wk low was due then too). using one of Hurst's tools, a commonality phasing model (which records prominent lows of a number of individual stocks) does reveal some prominent 80 wk/40 wk low clusters that did bottom in the mid january to very early feb time frame. the spread of individual stock 80/40 wk lows that show in that time period suggests that the 80/40 wk low on broad indexes was muted as some index component stocks bottomed and began rising while other index component stocks were still declining, canceling each others price effect on the averages.. labeling that area as a 80 and 40 wk cycle low was important to be able to forecast the end of july/early aug 07 as the next 4.5 yr cycle low. it is also a 80,40,20,10,and 5 wk cycle low. i'm expecting a nominal 10 wk low to bottom at any moment, and the next 20 and 10 wk low is due to bottom late in december.

once i can get some charts up on imageshack i'll post 'em.

[Woody]

Thanks Aire........always enjoy your Hurst Work and Analysis!