Quote from Mavrick99:
I apologize in advance for threadcrashing...but I came across this thread as i was looking for historical intraday currency data and the above recommendations fit the bill perfectly...thanks!
Great. The easy part is behind you now...
Quote from Mavrick99:
I'm looking to backtest a few strategies where specific bartimes have relevance. Therefore, i'm working on a script to adjust the time of Forexite's data to localtime, but am not sure how to handle daylight savings time (DST).
Forexite indicates their data is GMT+1. Friday closes are 2200 GMT+1 throughout the year.
Does anyone know how i relate these back to US times and take DST into effect? If NY is GMT-5, and during daylight savings is GMT-4, this implies during Standard time the currencies stop trading on Friday's at 4:00pm, but during DST trade until 5:00pm(2200-1-4)?
Can anyone tell me if this is accurate or am i missing something??
TIA,
Mav
Pragmatically speaking, your end result turns out to be correct, although it would be more accurate to state that Forexite's trading hours end at 22:00 GMT+1 every Friday year-round.
New York currency trading hours still unofficially end at 17:00 ET (whether EST or EDT) every Friday year-round, rather than at different times.
Both statements above are true regardless of whether DST is or is not in effect in New York. Like GMT itself, Forexite's historical data does not observe DST - an empirically verifiable fact, by comparing against other, known sources.
Therefore, in pseudo-code,
if DST = True, then NY Price [t] = Forexite Price [t - 5 hours];
if DST = False, then NY Price [t] = Forexite Price [t - 6 hours].
Since the US DST rules changed last year, you'll need to be careful with the exact DST dates and times each year for your backtesting or maybe even hard-code them. Here's one reliable source, out of many:
NY DST Dates 2000-2009