Hello,
Last night I was toying with the '33-liner' code and am having some issues trying to successfully implement this strategy. It currently returns a W/L of about 55-60% on my universe of stocks/timeframe but the system only generates decent returns if I choose 100% equity on a per stock basis.
What is the proper PosSizer to use on this default strategy out of the gate? What exactly is the priority loop at the bottom of the code signifying? I've changed the int i = 0 for the bottom loop without success as well.
Any information on how to interpret or utilize this strategy would be great, thanks!
Size:
Color:
User Guide > Reference > Data Panel > Position Size Control > Raw Profit Mode
And since you're using 100% sizing make sure you understand it how it works...
User Guide > Strategy Window > Backtesting Strategies > 100% of Equity Sizing
A first step to understanding a strategy is to run it in Raw Profit mode with a fixed dollar amount that's greater than the highest price of any of the stocks in the watchlist. This ensures that you'll get all historical trade candidates that were possible over the test period. Study that result - because any of those trades could be a "100% position" if you that sizing.
I can't speak to the 33-Liner specifically. Where did you get the code? I don't have it.
Size:
Color:
QUOTE:
I can't speak to the 33-Liner specifically. Where did you get the code? I don't have it.
I guess from a forum thread like this:
what performs best?I'll work on uploading it later today.
Size:
Color:
Size:
Color:
Thanks Cone and Eugene.
In the code it looks like it has a priority loop at the bottom, possibly up to 16 positions. If I use an equity size of 6.5% the strategy just doesn’t work so I was curious if I was interpreting it wrong. I’m just looking at this strategy as a starting point to potentially build upon.
As for the 100%... I ran it as portfolio mode and then utilized the Monte Carlo feature which should be very similar to raw profit. What’s interesting is that it has a 55/45 win rate and the winning amount % is slightly higher than the losing amount %. If I run it with anything less than 100% equity for multiple positions the results start to crumble. Thoughts?
Thanks again.
Size:
Color:
QUOTE:
Monte Carlo feature which should be very similar to raw profit
Not really. But if you're going to do that make sure to use the Same Date Scramble option.
If you're at 100% in a Port. Simulation, there's no room for more trades until the active position is exited. These n-liner scripts keep putting on more and more positions, which you will see in RP mode.
Size:
Color:
Thanks Cone, what pos sizer do you recommend trying with this script?
Size:
Color:
Re:
QUOTE:
Uploaded:
I experience two issues:
1. Clicking "33 Liner" in Chrome gives""Server Error in '/' Application. Object reference not set to an instance of an object." The link works in IE.
2. Following instructions in the target page downloads lots of strategies, but not "33 Liner", in "Counter-Trend" or any other category.
Size:
Color:
Fixed. Please retry.
Size:
Color:
Works now - thanks.
Size:
Color:
Good work as usual
thanks
Size:
Color:
So does anyone have success with this strategy using a position other than 100% equity? I can’t seem to find any combination of pos sizers or equity sizing that seems to be positive. I am new so maybe I am implementing it incorrectly. Thanks again.
Size:
Color: