Overcoming History Gaps, Downloading History, Daylight Saving (alpari-forex.com)

 

Hey,

I believe that all the issues are related so I've grouped them into one post.

First of all, I've noticed that if on a machine that MT4 didn't run for a while, when I start it I might get some gaps in the history. When I press the download button the history slightly changes and some other gaps might get created.
This is strange since I expect that the history will be downloaded from Alpari (alpari-forex.com) and will be the same one as logged live by MT4 when it was running.
What's going on here? How do I overcome the gaps problem and make sure that I will always have a history that matches that live one?

Second of all, I was surprised to find out that Alapri is using EET time, i.e. The time on server changes when the lock changes. As I understood, it happens with other brokers as well.
EA that depend on hour of the day, must take this into consideration, this is something that is manageable, but my real question is about history, especially a one that was downloaded using the download button
How can one know if the time in historical data that he is using is adequate with the real history?

Thanks.

P.S. Sometimes MT is very good, sometimes there are so many lame things about it...

 
fudge:

Hey,

I believe that all the issues are related so I've grouped them into one post.

First of all, I've noticed that if on a machine that MT4 didn't run for a while, when I start it I might get some gaps in the history. When I press the download button the history slightly changes and some other gaps might get created.
This is strange since I expect that the history will be downloaded from Alpari (alpari-forex.com) and will be the same one as logged live by MT4 when it was running.
What's going on here? How do I overcome the gaps problem and make sure that I will always have a history that matches that live one?

Keep your MT4 Terminal running 24/5 . . .

Second of all, I was surprised to find out that Alapri is using EET time, i.e. The time on server changes when the lock changes. As I understood, it happens with other brokers as well.

EA that depend on hour of the day, must take this into consideration, this is something that is manageable, but my real question is about history, especially a one that was downloaded using the download button
How can one know if the time in historical data that he is using is adequate with the real history?

History that you download from the History Center comes from MetaQuotes not from your Broker so may be on a different timezone to your Broker's data . . . don't mix your data, get it all from one source.
 
fudge:

...

How do I overcome the gaps problem and make sure that I will always have a history that matches that live one?


It happens. I did not spent much time on it, but I tried this.

1. alternative : Removing all the .hst files fixes the gap issue. But some history gets lost.

2. alternative: In the history centre, cycle through the ticker timeframes and select&delete all the bars except the oldest. It may create a gap just in the beginning, but for me it loaded more bars the the alt. 1

Then it is good to cycle through the timeframes and right-click the "Refresh" option on every of them.

fudge:

..

How can one know if the time in historical data that he is using is adequate with the real history?

Good question. I believe historical data are under full control of the particular broker. You may ask them if they edit them, but I suppose that they do not bother to change them once recorded.

 
fudge:

Hey,

I believe that all the issues are related so I've grouped them into one post.

First of all, I've noticed that if on a machine that MT4 didn't run for a while, when I start it I might get some gaps in the history. When I press the download button the history slightly changes and some other gaps might get created.
This is strange since I expect that the history will be downloaded from Alpari (alpari-forex.com) and will be the same one as logged live by MT4 when it was running.
What's going on here? How do I overcome the gaps problem and make sure that I will always have a history that matches that live one?

Second of all, I was surprised to find out that Alapri is using EET time, i.e. The time on server changes when the lock changes. As I understood, it happens with other brokers as well.
EA that depend on hour of the day, must take this into consideration, this is something that is manageable, but my real question is about history, especially a one that was downloaded using the download button
How can one know if the time in historical data that he is using is adequate with the real history?

Thanks.

P.S. Sometimes MT is very good, sometimes there are so many lame things about it...

  1. I find it better to have a M1 chart open for each currency I wish to maintain a complete history. Using the M1 data, I can better use the strategy tester and I can also convert M1 data into whatever higher timeframe I desire.
  2. The broker you mentioned changed the time zone for its MT4 servers from CET (UTC+1) to EET (UTC+2) in April of 2012. See broker's announcement here. The broker also observes daylight saving time, so CET becomes CEST (UTC+2) and EET becomes EEST (UTC+3) during daylight saving time. However, it is pretty easy for an EA to adjust for daylight saving time. See my posts (with example code) here for European DST and for American DST. Also remember that any history for that broker which precedes the time zone change is probably in the old time zone (CET).

Reason: