Poll: How often do you reoptimize ?

In your trading strategy how often do you reoptimize your settings to have the best fit on current market conditions ?

 

In my case (equity options trading) optimization means to refresh the watchlist. This task requires scanning (ie. filtering), ranking, and time-consuming manual analysis (ie. manual verification) of those top ranked...
Takes in my case about 4 hrs each time, incl. downloading of the data of all tickers (this alone takes about 1.5 hrs )... So doing this only once every 2 weeks or so... b/c costly in terms of money, work and time...

 

it should be centered around deviation of current and historical pattern reliability. not some man made time interval.

 

Okay, but to be concrete, when did you last time reoptimize your strategy ?

 

first when picking parameters i don't choose best results i choose something in the middle of the pack. i know that by dumbing down the performance i can gain long term stability. then only when the model deviates beyond it's historical drawdown. still the system will return half it's losses back. around that comeback i will reoptimize and see what's up. m

 

in 2011 there is discussion about this subject

 

I'm not sure that's really optimisation, we're talking about some kind of parameter fitting process I think.

I run a daily optimisation to determine what trades I should do, but the parameters that control the forecasts which dictate those trades are never optimised. So I answered 'never'. The parameters are based on 50 years of data. Having another year of data is irrelevant.

Most people optimise too much and too often. Even HFT guys only optimise about every 3 - 6 months.

A lot of this is to do with people not designing strategies that automatically cope with changes to eg volatility, price levels and so on. If you don't scale your forecasts according to vol estimates, you'd have to reoptimise every time volatility changes.

GAT

 

Good Morning TrAndy2022,

My eyes LOVE and MASTER every market conditions the ES futures market brings my way.

I will never reoptimize nothing.

 

%%
Good ; TM did it well in the 1970's, just in time inventory, great co, but i never do that.
But related to that\ time about SEPT sells month are very common in stocks\ETFs\Apples.......
I constantly do research + trend study, but its divided in bull + bear markets more than a man made time . So typical poll= not enough choices , but good question

 

So, then mine is similar: the parameters are constant, don't change; only the watchlist content changes.
I now have changed my vote to "never" from "every 2 weeks".

Yes, agree. Optimization should also include and auto-optimize such of the processes.