Backtesting/Optimization - page 14
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
M1 data older than 2004/06/14
Where can I find reliable M1 data for the major pairs that is older than 2004/06/14? I'd like to find at least 4 years of data. I realize I may have to purchase it, but that's okay if the price is reasonable.
Where can I find reliable M1 data for the major pairs that is older than 2004/06/14? I'd like to find at least 4 years of data. I realize I may have to purchase it, but that's okay if the price is reasonable.
It is here https://www.mql5.com/en/forum/174887
And some data is here as well https://www.mql5.com/en/forum
About how to install and backtest you may read Codersguru article and this thread.
Question about the strategy tester. My EA does very well on the Alpari data. But once I use the 2001-2005 data I found, it goes choppy. I'm starting to question the quality of the data I'm using. Is this normal that data from one source is of bad quality? What's the consensus on the Alpari data?
Modelling quality is 87-89%, only once have I seen 90%.
Question about the strategy tester. My EA does very well on the Alpari data. But once I use the 2001-2005 data I found, it goes choppy. I'm starting to question the quality of the data I'm using. Is this normal that data from one source is of bad quality? What's the consensus on the Alpari data? Modelling quality is 87-89%, only once have I seen 90%.
Alpari data should be ok. I always have 90%.
Concerning the gaps in data so there is some script to find the gaps and the other script to fill the gaps: it is in MetaEditor, library and so on.
Backtesting Results
Ok, I am experiencing a very wierd problem. I will run a backtest and get one set of results, but I close the program down and reload, and now I get completely different results. Anyone have any ideas what the deal is?
Hi Ducati,
I think I know whats happening, the only way to get a reliable backtest in metatrader is to make sure 1) you have 90% modelling quality and 2) don't open an account eg: always use metatrader offline, this is very important and not explained anywhere when you open an account you overwrite some history data thus changing results. I recommend a seperate MT4 install just for testing and don't open an account just offline. Regards, Lee
Ok, I am experiencing a very wierd problem. I will run a backtest and get one set of results, but I close the program down and reload, and now I get completely different results. Anyone have any ideas what the deal is?
it is good to have 2 MT4 platforms installed in your computer, one for demo/live accounts (forward test therefore), another for backtesting, which you will only use offline, importing its data from other history databanks...
Oh, I didn't know about the offline thingy. I was wondering why I started with 90% modelingquality and now its dropped to like 85%.
Thanks!
Oh, I didn't know about the offline thingy. I was wondering why I started with 90% modelingquality and now its dropped to like 85%. Thanks!
Ducati,
which version of Metatrader are you using (Help / About...)?
Was your MetaTrader automatically upgraded when you restarted?
I experience similar behavior since upgrading build 198 to build 200 and am unable to get anything backtested correctly anymore.
Build 200 uses an internal download mechanism for history data.
Origin of this data is unclear to me.
I used to import and backtest with Alpari history data.
This always backtested the same way as my life trades with NortFinance.
Herbert,
Yes, it did change immediately after build 200 was downloaded! I completely forgot that it downloaded a new build.
I also just downloaded a brand new version of metatrader from the metatrader website and it is version 4 build 200 and it will not let me import the historical data from Alpari. I select the file, but nothing happens. This sucks!