Hi,
I have a strategy I'd like to implement - and I'd like some advice before I get too far down the road.
I have a scan I run daily in AmiBroker that completes a setup scan that will yield anywhere from 0-350ish potential underlying candidates for the next day.
I then need to enter positions in a series of scale in limit orders, exit either MOC or when a position average profit target has been reached (i.e. 10% from position average price) - which changes/updates as fills are achieved thereby changing the average price.
The maximum number of positions will be < 15 or so, the actual number being determined by the number of underlying candidates hitting the initial limit order but in no case more than 15.
Sooo......
My quote provider will let me subscribe to up to 500 symbols.....so that's no problem.
I can export a list of underlying candidates to a CSV file that I could then read into NT, so that's no problem.
I have the algorithm well defined - so that's no problem. I can probably get by with placing the initial order on initialization and placing/managing additional orders on execution.
BUT
Is it a good idea to ask NT to track 300 symbols and then potentially manage orders for 15 underlyings in a SINGLE strategy? Note that I have 2 accounts I'd like to run this on - so I'd be running 2 of these bad boys at the same time. This is my preferred method of accomplishing this as I would be able to completely automate my strategy.
A second approach would be to create a watch list outside of NT, and create a single stock strategy for each candidate that could be started as the issue got close to the first limit order. In this scenario, I'd have potentially 30 of these strategies running for different symbols/accounts. The drawback here is I'd have to manually watch my watch list and start strategies by hand.
Thoughts???
John
I have a strategy I'd like to implement - and I'd like some advice before I get too far down the road.
I have a scan I run daily in AmiBroker that completes a setup scan that will yield anywhere from 0-350ish potential underlying candidates for the next day.
I then need to enter positions in a series of scale in limit orders, exit either MOC or when a position average profit target has been reached (i.e. 10% from position average price) - which changes/updates as fills are achieved thereby changing the average price.
The maximum number of positions will be < 15 or so, the actual number being determined by the number of underlying candidates hitting the initial limit order but in no case more than 15.
Sooo......
My quote provider will let me subscribe to up to 500 symbols.....so that's no problem.
I can export a list of underlying candidates to a CSV file that I could then read into NT, so that's no problem.
I have the algorithm well defined - so that's no problem. I can probably get by with placing the initial order on initialization and placing/managing additional orders on execution.
BUT
Is it a good idea to ask NT to track 300 symbols and then potentially manage orders for 15 underlyings in a SINGLE strategy? Note that I have 2 accounts I'd like to run this on - so I'd be running 2 of these bad boys at the same time. This is my preferred method of accomplishing this as I would be able to completely automate my strategy.
A second approach would be to create a watch list outside of NT, and create a single stock strategy for each candidate that could be started as the issue got close to the first limit order. In this scenario, I'd have potentially 30 of these strategies running for different symbols/accounts. The drawback here is I'd have to manually watch my watch list and start strategies by hand.
Thoughts???
John
Comment