99% modelling quality but with a red bar - page 4
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Re: the red bar, are you 100% certain that you are copying over all the hst files, from M1 to MN1 ?
For example,
GBPUSD1.hst
GBPUSD5.hst
GBPUSD15.hst
GBPUSD30.hst
GBPUSD60.hst
GBPUSD240.hst
GBPUSD1440.hst
GBPUSD10080.hst
GBPUSD43200.hst
Yes, after downloading the data, the first thing that I done was run the script to generate all the hst files and the h1 fxt file.
I placed all the hst files into the metatrader/history folder and fxt file into metatrader/tester/history.
I thought the backtester just uses the FXT file?
Also confusion over the GMT offset. Dukascopy data is GMT+0. So I assume that I can backtest with the GMT offset set to 0, despite my broker currently being GMT+2 and following DST? I am guessing the tester data is completely independant from the broker, thus 17pm on my broker would still be 17pm GMT+0.
I'm confused about GMT vs Data-Time myself :). However, I believe, if you're using Dukascopy's data. Then Your broker while you perform a backtest should be taught of as Dukascopy. If your data is from Forexite, then Your broker while you perform a backtest should be taught of as Forexite. If you wanna place an order within the back-tester @-X-Gmt you'll have to know the Brokers Time Shift and If and When they adjust for DST.
My confusion about this is THE primary reason why I haven't created any system which relies on Universal Time, it's also one of the reasons I haven't bothered to download Tick data.
Take a look at WHRorder's codes within this thread. It gives pretty good examples on how to use Shifts in back-tester and DLLs to get Universal Time while running Live.
See as back-testing is an estimate of how the system performed. I think you're taking it too seriously. The usual recommendation for Tick-data are for systems which places well over 200 trades within 5 years. Also recommended for systems which usually close within the same minute bar upon which it was opened. Mostly systems looking for single digits profits would fit into this category.
If you truly believe your system is a scalper worthy of the headache, then I say it'll also be worthy of a prolong forward Live test. But with that 200 trades in 5-years, I don't think it is. This is just my Opinions, by-the-way.
I thought the backtester just uses the FXT file?
My system is a scalper. It can close trades in seconds or minutes. Hence, the need for accurate tick data. The GMT offset I have confirmed should be left at 0 as the data is GMT.
The only things bothering me are the red modelling bar (somebody in this thread stated it was because there was no modelling needed) and the losses when using normal history data that dont occur with the Dukascopy data.