Spuddy
-Interested User-
Posts: 4
Joined: Aug 1, 2004
|
Posted: Aug 1, 2004 02:15 PM
Msg. 1 of 2
This is my first attempt at retreiving IQFeed's historical minute data via TCP/IP and I noticed that the data appears to be very unclean at times. Indices such as COMPX.X appear to be fine, but after retrieving 100 days of minute data for QQQ, there is usually at least one bad minute bar each day from what I assume is unfiltered Tick data being used to generate the minute data. Other stocks' minute data is similarly dirty. Are there any plans for DTN to filter out the bad ticks before it's used to generate the historical data ? I don't mind being delivered unfiltered Tick data but I'd like to be given accurate historical data. I surpose I could retrieve the tick data and do the filtering myself but that sort of defeats the point of being able to access historical minute data. Only 8 days of Tick data are supplied too and 120 days of dirty minute data is available. I can always filter my minute data and drop the bad bars but now I'm wondering how accurate longer time frames are such as 60 minute and daily data is if bad Ticks aren't being filtered out at DTN's end ?
Also, when I retrieve a days worth of minute data for a particular stock, I've noticed that I'm not always delivered data for every single minute in the trading day. Is this because there were no trades during the minutes that IQConnect skips or shall I start looking for a bug in my code ?
Thanks
|
DTN_Jay_Froscheiser
-VP, Product Operations-
Posts: 1746
Joined: May 3, 2004
DTN IQFeed/DTN.IQ/DTN NxCore
|
Posted: Aug 2, 2004 11:16 AM
Msg. 2 of 2
We are working on some filtering mechanisms for the intraday data. In the mean time, if you see bad ticks, email them to customer support and we will work to get the corrected.
If an issue didn't trade during the interval you chose, you will not see a bar (since you can't have an open/high/low/close if there were no trades during the period).
Jay Froscheiser DTN Market Access, LLC.
|