Welcome to NexusFi: the best trading community on the planet, with over 150,000 members Sign Up Now for Free
Genuine reviews from real traders, not fake reviews from stealth vendors
Quality education from leading professional traders
We are a friendly, helpful, and positive community
We do not tolerate rude behavior, trolling, or vendors advertising in posts
We are here to help, just let us know what you need
You'll need to register in order to view the content of the threads and start contributing to our community. It's free for basic access, or support us by becoming an Elite Member -- see if you qualify for a discount below.
-- Big Mike, Site Administrator
(If you already have an account, login at the top of the page)
Rick, this is why Van's concept of R is so important. If used correctly it normalizes the system to any account size (provided the account is of adequate trading size to begin with).
I have been doing some serious thinking about expectancy and the routines I wrote. I think my fundamental theory is flawed. The expectancy and expactancy score routines do the math correctly, but they utilize the NT Optimizer to operate. The optimizer (or backtesting if you will) is a serial analysis. That is you start in trade 1 long and follow that trade until you hit profit or you stop out. Then you look for trade 2 which let's say is a short and follow that one along until profit and or stop is reached and so on.
I think the way it SHOULD be done is in parallel. That is, you take every possible trade (long or short) and determine what results you would have achieved with your strategy and then you optimize to that.
I don't believe you can get their with Ninja. So where does that leave us? A different kind of optimizer...[on sale now for only $49.95...but if you act within the next 30 minutes, we will send you not one, but two optimizers...but wait there's more...we also have these amazing ginsue knives that we will give to the first 40 callers...so call now at BR-549!]
Well different optimizers are possible, but extremely complex and based on my last experience a couple years ago they either didn't work with NT7 or worked very poorly.
Many folks like the MoGo optimizer better than PH Genetic. It certainly has more features. I personally use the PH Genetic but decided to port the MoGo to 7 as an exercise to deepen my knowledge of programming. Attached is the NT7 version. It is completely …
I have read the expectancy's threads, thanks for the optimalization script. My question , is it possible to add custom fields to the backtest result tab (to the summary grid) ? For example if it is possible I would like to edit that tab , and add it an expectancy field.
Maybe said another way is that a "positive expectancy system" should have an equity curve that is smooth and upward sloping.
An older now unavailable product called "The Grail" provided a nice means of quickly being able to see the resultant equity curve for each of the system tests. I personally found this a good means of identifying the systems that were worth evaluating further.
My thought is if would be possible to create an fitness function that identify systems with a strong upwardly sloping equity curve and was very smooth.
Maybe ... <Slope of Linear Reqression of Equity Curve> * <function that calculates "linearness" Equity Curve>
Maybe this is the end result of some of these advanced fitness algorithms ?
"More importantly I think after doing some studying that if you want to express the expectancy in terms of multiple R you can only do that when the risk per trade is a fixed amount. This is also obvious. This is also flawed. You will not risk the same with a high beta stock as with a low beta one.
Too much hype in this industry... "
Some people adjust for the beta using ATR as their stop. This allows you to trade consistently over each market type.