Topic: Timezone of backtest data


i have a question about timezone of backtest data.

I use tickstory to get 99% Quality backtest data. For exporting the data what timezone i need to use? Did i need to use my own pc time? Did i need to use the server time from broker?

What ea will do? I generate strategies with e.g my timezone of my broker i run backtest and get same reults as eastudio give me out. What is wehn i use the Eas to another broker. Will they work or will the damaged the balance?

Re: Timezone of backtest data

what about older generated strategies. did i have to change a timeoffset or somehing when broker change gmt time?

Re: Timezone of backtest data

hmm nobody an idea.

Okay another timezone question.

When i like to generate strategies of DAX or and Dow Jones. Did i have to set in Settings trading time like for dax.

9:00 - 17:30 (DAX)
15:30 - 22:00 (DOW JONES)

Did i have to change times also for gold?

Re: Timezone of backtest data

is there nobody who can tell me something with the usage of the backtesting time?

Re: Timezone of backtest data

If you're not using your broker's data, it's still useful to use the same gmt time as your broker (forget your local time). Practical experience shows that you will have performance issues if development data origin is different to following trading data, the more robust the strat, the less performance difference there is, theoretically speaking.

How's your dax data looking? Real bars from 9-17 and in-between just last known price?

Re: Timezone of backtest data

but what about the time..

if i use dukascopy data. and i use different broker data for backtesting so the time will be important?

did i only have to use brokers time? Cause i want to use dukascopy data for eastudio. but will also test my eas with different brokers data. so if they have different times i have to switch it all the time?

So if i want to use this ea in different brokers than the ea will not work on broker b or c cause the time is other?

It it will be so . So it i think there must be something a gmt offset inside ?

Re: Timezone of backtest data

When changing the broker, run a quick MT backtest to see the performance. It may not be absolutely the same but may be similar.

The timezone matters when you use a sort of time limitations or periods greater than H1.

Anyway, the best is to use data from the same timezone in order to be easier to compare the orders.

Re: Timezone of backtest data

so isnt it like this.

When i use dukascopy data and use it for backtesting with my local timezone. So i get the same backtest as in eastudio.

Now when i use broker A with a timeshift of 2 hours. Than e.g the 1hr candle is 2 hours shift from my localtime and dukascopy eastudio data.
So the ea calculate from its calculations with the shiftet candle. So it must be the same as normal. So the time is irrelevant.?

Re: Timezone of backtest data

The candle form of H1 and lower periods will be the same. If you don't use Trading Session or other time limitations, the strategy will not depend on the time zone. It will simply not be possible to distinguish it.

However, other factors may cause a difference. The bar ranges may be very different. The volume also will differ for sure.