Just got a very general question about indicators of pretty much any sort, when plotted on, for example, 5 min charts over a series of days.
Lets take the ES 06-08 contract over the past two weeks as an example. 5 min bars plotted over those two weeks. I have defined the session start and end times to 14:00 and 21:15 respectively GMT (London UK time). Much like looking at stock prices (which are obviously closed overnight) I'm only interested in the main trading session.
Now, much like plotting similar data for stocks, I get, obviously, some decent overnight moves between session breaks. These large 'overnight' moves obviously have a marked effect on any indicators that I'm plotting on the 'continuous data'... that is 2 weeks worth of 5 min data. If I'm plotting an SMA(14) and the ES 06-08 contract has rallied 40 ticks overnight then clearly my SMA will be skewed at the start of the day.
I guess what I'm asking is, how do day traders who are plotting many days worth of data (and backtesting) deal with this issue. As far I can see there is only one way to deal with this...
1. I put code into every indicator I use on data like this to make it only plot the inidactor using data from the current session. i.e. the indicator will effectively have a 'session break' itself.
Am I missing something...?? If not, what would be a real bonus is if NT could add a flag that if set to true, only data from the current session was used to calculate an indicator.
Comment