Jan

12

 I asked Brett one more time to quantify some of his multiple classifications and descriptions of cycles and advance declines et al and he responded: "Very nice quanatification of cycles with admirable transparancy of track record fromm @stockspotter. Brett is a Dr of great respect and is adverse to mumbo in all forms. But someone should look at stockspotter to see if it passes the usual scientific standards.

Phil McDonnell writes:

I first came across John Ehlers in the early 1990s. He and I traded his MESA program for my Option Trader software. MESA stands for Maximum Entropy Spectral Analysis. I had earlier written a Fast Fourier Transform (FFT) program which is similar to MESA. FFT constructs a set of multiple cycles over 32, 64 or 128 days to 'fit' the recent history of the market. The problem is that cycles evolve with time. MESA attempts to fix that by taking a shorter term average cycle than FFT.

My experience then was that the software seemed to have some promise but it was not perfect. It offered a look-ahead graph of where the market would go and that was how you picked potential winners.

A quick look at the stockspotter.com site showed a track record of 'closed trades" for the last quarter of 2016. I glanced at the first 10 pages of trades. Out of this sub-sample of 140 trades very few were losers. On the face of it this seems to be a very successful system.

But there is a catch. The trade record shows only a few days (3-6) on the first page. But later pages show a clear trend of more days in a trade (~5-9) as it goes on. This may be evidence of some sort of 'first profitable close' exit strategy. Thus there may be a larger number of deferred losers hidden in the undisclosed current trades which were not revealed.

As it stands the data cannot really be analyzed scientifically. However that does not mean that there is not something interesting here.


Comments

Name

Email

Website

Speak your mind

Archives

Resources & Links

Search