Hurst is quite famous for writing a book. It's called The Profit Magic of Stock Transaction Timing, by J M Hurst. It's available from Global Investor, etc.
I have the book, and it's quite mathematical but it does give you all the details for working out cycles. From memory, it's not very expensive, so perhaps you may wish to read the book before you subscribe to the service. It was first published in 1970, so is quite an old-fashioned book, if you know what I mean.
I subscribed to & papertraded TRM for 3 months. As with most technical analysis the historic charts seem to work very well, but when using the system in realtime flaws become apparent. In particular their charts are "adaptive", so that for one or two days their "signal line" (on the medium term sigma band charts, or shorter term signal band charts) may show an entry signal, but after factoring in the next day or two`s data that signal can often disappear. Also, when prices cross over the +/- 2 or 3 sigma band lines (which are standard deviations from a long term trend line) they can often stay there for a very long time or continue in their prior trend, rather than reversing within days as implied by TRM (their website confidently states a 98% or 99% probability of a reversal at these lines, which may be mathematically correct, but it doesn`t mean the reversal will occur right away). The "cycle decomposition charts" & "curvilinear envelopes" are interesting, supposedly for gauging the current trend, but in practice of limited use, due again to the adaptive nature of the indicators. I did observe the sigma bands working very well on a few occasions, but in my experience the signals given by the system weren`t reliable enough for me to use them as a stand-alone trading system, and their monthly fees were too high to keep subscribing as a secondary system to help confirm other trade signals. TRM allow you to try out their system for only $2.95 for 14 days, so I suggest you give it a try. If their testimonials are to be believed many of their subscribers love the system. The guru who developed the system from Hurst`s work, James Maggio, is very helpful if you have any questions.
Thanks for the feedback fellas.
Tradertim,you bring up some interesting points,you say when prices get to the 2nd or 3rd band it can take a long time before they reverse.How long is a"long time"?If it is,say,a month or so then it would be suitable for doing credit spreads,particularly if the prices do indeed reverse 98% of the time.Anyway I have browsed their site and find it a bit curious that everyone who has sent a testimonial have also given their email address.Interesting!!!
It`s a while since I subscribed to TRM, but I seem to remember prices on US index charts staying above their +2/3 sigma bands for about 3 months (on longer term sigma band charts) during the period when there was a strong upward trend in force last year. Even when the price drops below these bands there is of course no guarantee that they`ll retrace back to any great degree, especially if there is a strong underlying long term trend in force. On the shorter term signal band charts, subject to the previously mentioned caveat about the adaptive nature of the system, prices seemed to react to the sigma lines in a more immediate way, but by the time you`d observed this (on end of day charts) and the crossover of the signal line (like a moving average crossover), the larger part of the move had often already occured.
It may be a coincidence, but today I received an e-mail from TRM warning about alternative providers of Sigma Bands. It says they are based on "polynomial regression and not critical analysis" and that others are using "pretty questionable ideas concerning momentum"...... and no, I don`t know what that means either,
but I thought you should know. TRM say that they`re consulting their lawyers, so it looks like Ben Dover is about to get shafted.
Once again great feedback.
By chance I came across another website optionsedge.com and I think they run along the same lines as TRM.The person who runs this site is one Travis Mijat who supplied one of the testimonials on the TRM site.
Instead of sigma bands they have a price drawn on a chart which,according to them,is the limit the price will go to before it reverses thus giving option selling opportunities.They gave me a 2 wk trial and I have noted about 50 stocks that should expire away from this price come Feb expiration day.I'll post again then and let you know how they did.
BTW they give these prices for US UK and AUS stocks as well as Indices.
Another website found, seems to be run by the Trading Academy already discussed on these boards. It's called Trendsignal and by looking at some of their chart samples they seem to be using something similar to Sigma Bands. These bands may be offset moving averages rather than standard deviation, cant tell for sure http://www.trendsignal.co.uk/show_me_more.asp
I didn't subscribe, but got a free 2-week trial. I agree that the sigma bands weren't reactive enough. Worse, there is no law to say that they may not go to the 4th or 5th sigma after a few months. So during my free trial I decided to filter the signals thru the Fibonacci levels. If a Fibonacci cluster occurred at the 2nd. or 3rd. sigma band, I expected a quick reaction from those areas. I also added a simple 2-bar reversal pattern as the trigger to trade. As luck would have it, I got an alert for the cattles, having reached sigma 3. I have a friend who owns the NexGen T-3 and he told me there was a Fibo cluster nearby. I then waited for the 2-bar chart reversal to appear before entering the trade. The rest is history, thanks to mad cows. Before subscribing I discovered that Jan Aarps was selling the sigma indicators for $695 which represented 4 months of subscriptions. So what do I do? Buy 4 months of milk or buy the cow? Tradertim was correct when he said that the other vendors might be selling "polynomial" maths. But this isn't a crime, and in any case I was combining those sigma alerts with Fibonacci clusters and chart pattern. Believe it or not. I didn't buy the sigma indicators nor subscribe to TRM. Instead I bought an interesting set of similiar indicators from the land of Oz....The Voodoo Chooks by Bob Wendell in Perth. (A$800). I will not try to influence you with any rave report. Just check out his website and see if the Voodoo Chooks suit your trading style.
By the way, Travis Mijat is also from Perth. He has nothing to do with sigma bands. His forte is with Elliott waves, having worked some years ago with Richard Sanwell supporting the Elliott Wave Analyser. Now Richard offers an Elliott Certification Course for $4000 using almost identical methods of analysis as Travis.
Please be careful when you dabble with statistics if you don't fully understand their derivations. For example, if a trend has been established using Linear Regression and you plot lines at 1 or 2 standard deviations (sigma) from the regression line, you can make statements about the percentage chances of a price going outside the standard deviations. The mistake people make is to assume that there is a high probability of the price returning towards the central line. As stated by others, this is only true if the original trend is still valid. If the trend that generated the standard deviations is broken, the price may never recover. I know this to my cost having bought Marconi at 2 sigma below an upward trend line, assuming it would recover. It didn't, ever!
I agree it would be foolish to rely upon regression to the mean if this is regarded as a fixed point. If a statistical method is to be used then it is important to understand that the probability of regression is subject to the referenced mean, it is therefore as dynamic as that mean. Given this you need to understand the nature of the reference point, for example a moving average is far more volatile and subject to manipulation than say a Volume Weighted Average Price.