Data issue for GBPUSD on Brexit

[Milonnewday]

Hi,

I found out there is a mismatch in DateTime and the OHLC daily in GBPUSD.
There was the lowest exchange rate of GBPUSD on 20161007 since Brexit. However, the data downloaded through Tickstory indicated the lowest exchange rate is on 20161006.
The data downloaded through Tickstory for the date below:
1.20161006 0 1.27521000000000 1.27543000000000 1.19198000000000 1.23878000000000 533590.747602053


I hope anyone can look at this issue, and hopefully, it can be solved soon.

[tickstory]

Hi Milonnewday,

I'm not sure of what time-zone or broker you are comparing with, however be reminded that FX exchanges are decentralized and can therefore have differing prices depending on trading activity in their liquidity pool.
If you compare with this data source, for example:

https://www.macrotrends.net/2549/pound- ... ical-chart

You will see that a low actually occurred on the 11th of October 2016 with another even lower-low occurring on Jan 16th 2017.

The data you see is a reflection of what occurred on the platform at the time and there seems to be no reason to suspect a fault.

Hope this helps.

[Milonnewday]

Thanks for your reply.

Both data I compared in UTC. I plot two charts for this issue which I think the DateTime mismatched the OHLC in one day lag.
The first chart file came from Tickstory, which you see the spike candle found on 06/10/2016.
The second chart file came from Dukascopy, which the spike candle found on 07/10/2016.

I also compared this spike candle with Tradingview crossing several CFD brokers, and all of the data sources showed me that this long spike came from 07/10/2016
The link from Tradingview:
https://www.tradingview.com/chart/R5uoZkjS/

[tickstory]

Hi Milonnewday,

Thanks for the further info. We can't comment on the differences between TradingView and Dukascopy since of course we don't know where the former sources their data.
As for the Dukascopy data, we did download both the tick data and bar data from their historical data page here:

https://www.dukascopy.com/swiss/english ... istorical/

and noted that the tick and candle data correctly matches the bars that Tickstory has created (The low price was hit at approximately 23:09 GMT on 6th October 2016).

We can only assume therefore that there is either a discrepancy within the JForex platform (time-zone glitch?) or the two have different sets of data. We have had reports that Dukascopy candle data is not as complete as their tick data, so I'm not sure if that's the cause. As far as Tickstory is concerned, this should not affect the output as Tickstory builds all its own candles from the underlying tick data.

You may get some traction with Dukascopy support if you could point the differences in the two sets of charts but unfortunately we can't do anything on our side to alter their data.

Hope this helps.