- ago
I have developed several strategies. I am experimenting with just trading in the present with more profitable strategies during the short term. Similar to looking at "trading the equity curve".

So what I have been doing is I put 6 strategies into the MetaStrategy. However, on the strategies I am not currently trading, I put a Portfolio Weight of 0 for ones I am not currently trading, so that I only get signals on currently traded strategy.

As a test, I have one strategy with a portfolio weight of 1 and the other 5 strategies have a portfolio weight of 0.

I have found that with this setup, the MetaStrategy still runs significantly longer.

With out removing the 5 not traded strategies, is there a way to improve performance?

Thank you,
Larry
0
415
6 Replies

Reply

Bookmark

Sort
Glitch8
 ( 11.29% )
- ago
#1
Just playing around here with my MetaStrategy I have not noticed any slow down at all setting the weight to zero for multiple strategies. I wonder why you're seeing that.
0
Cone8
 ( 7.89% )
- ago
#2
When testing many symbols, usually significant slowdowns are caused by one or more of the enabled Event Providers. Disable any Event Provider that you don't need.
0
- ago
#3
I just ran a MetaStrategy with 6 strategies each having 0 weight. None had any Portfolio weight > 0. Took a bit of time to run. So I guess there is overhead in using a 0 Portfolio Weight strategy in a MetaStrategy. So my guess is each 0 Portfolio Weight strategy will use some processing time. Is this correct?

PS. I only have 2 event providers (Wealthdata and Yahoo Finance).

But for the 6 strategies I do have several datasets, with some of them large. Such as Russel 2000 and S&P 500.

Thank you,
Larry
0
Glitch8
 ( 11.29% )
- ago
#4
No, like I said I tested setting different ones to zero in my MetaStrategy and it had no impact. Why would you set ALL of them to zero?
0
- ago
#5
Glitch,

I set them to 0, so I can see if there is any overhead when setting to 0. So for me, there was.

My goal is to have the strategies have equal weight when they are making money so I will trade them.

When some of them are not doing well, just set the weight to 0 on those in the MetaStrategy. Also, this way, I don't see signals in the single MetaStrategy that I use for trading signals for strategies that are not currently traded.

The idea is I only put weight on those that are presently making money in the equity curve.

I am using a single MetaStrategy as a universe of my trading strategy and will turn them on when they are making money and assign 0 weight when they are not making money based on the equity curve.

This seems easier than adding and deleting strategies in my MetaStrategy when they are not making money.

Is this reasonable?

Thank you,
Larry
0
Glitch8
 ( 11.29% )
- ago
#6
Sounds reasonable to me.

I was asking why you were setting ALL of them to zero at the same time. If you weren't doing that, please excuse the question.
0

Reply

Bookmark

Sort