Hi, developing an automated trading system in NT.

I have tested multiple variables independently. Now I am looking to create a ranking system, where each variable is assigned a ranking depending on its optimum values. I would then add up all the ranking factors to come up with a decision to take a position and/or the amount of capital/risk I am willing to undertake.

Can someone please point me to a reference or provide some help as to how you go about doing this. This is a general question and not necessary NT related, but I do use NT.

steven2 

Hi, developing an automated trading system in NT. I have tested multiple variables independently. Now I am looking to create a ranking system, where each variable is assigned a ranking depending on its optimum values. I would then add up all the ranking factors to come up with a decision to take a position and/or the amount of capital/risk I am willing to undertake. Can someone please point me to a reference or provide some help as to how you go about doing this. This is a general question and not necessary NT related, but I do use NT.

@steven2,

Are you using the built in Optimization features on NT7? In a sense, this will optimize your strategy based upon input parameters and selected measures.

Hi NJAMC, yes, one could use the optimization feature. Which I do.

When you do that you come up with this huge matrix (depending on your variables), and then I think if you optimize that, it would curve-it. I would rather optimize on a range, meaning that I would check each variable separately, and then chart that variable to see if it has a smooth curve across multiple values, and no spikes.

I would then like to come up with the ranking data, that would work across a range of values.

If you just use the optimizer, please advise how you go about assigning optimized values, and not curve fitting too tightly.