Topic: 1m strategy and week close

Hello everyone...

I wanted to get people opinion on something .  As we use fab to find good strategies we keep the ones it produces the best back test. 

Let's pretend that we have a few good 1m strategies on eurusd all using various closing techniques (cgi, moving average,etc). 

As we forward test we eventually will come to times with the EA opens positions and the closing technique has not taken place before the end of the week.  If we do nothing the positions will remain open over the weekend and by the time the markets open again the price could have change quite a bit cause a loss.

In theory should we just ignore these losses ?  If the strategy fab devised did not include a week close / week close 2 technique then the backrest fsb performed also came across losses (and wins) as a result of keeping positions open over the weekend.

I feel the need to close all positions before the end of the week but that is just emotion.  If I stick to what fsb came up with I should leave everything in auto pilot. 

Thoughts ?

Re: 1m strategy and week close

Beyond just the normal gap there's another one that's not as obvious from a strategy generation standpoint.  As liquidity dries up at the end of the day on Friday the spread can sometime become gigantic in some pairs in the last few minutes before weekend close as market makers start pulling their orders from the books.  The generator probably considers the gap in the backtesting but as far as I know (other than in monte carlo) it uses fixed spreads so that might not get factored in.

Depending on the strategy just closing existing might not be optimal, might be good to avoid opening new ones at some point as well as some strategies can sit negative for a bit before swinging.

Honestly haven't tried to compensate for the gap/increased spread yet in fsb but it's something I've pondered myself.