I wrote a quick strategy to log price data on m1. I then waited a week and ran the strategy in the backtester on m1.
So, this means I have data collected in Live and in Backtest for m1 timeframe.
When I compare this data, it sometimes is different by small amounts. I'm curious as to why this difference occurs between backtesting and live price data?
For example here is one bar from start of last week (times in UTC) showing
EURUSD Open, Close, High, Low prices, and the last column is volume.
- Code: Select all
Live:
18/05/2014 23:19:00 1.36959 1.36951 1.36959 1.36948 33
18/05/2014 23:20:00 1.36951 1.3695 1.36951 1.36949 2
18/05/2014 23:21:00 1.3695 1.36952 1.36954 1.36949 18
Backtesting:
18/05/2014 23:19:00 1.36954 1.36951 1.36956 1.36947 33
18/05/2014 23:20:00 1.36951 1.36949 1.36951 1.36949 3
18/05/2014 23:21:00 1.36949 1.36949 1.36956 1.36949 20
You can see that there are slight differences. Why are there differences in these prices?
The only thing I can think of may be that the minute boundaries are different between my machine and the backtesting? I.e., the sampling is done differently.
Any ideas?
Thanks,
MooMooFX