So I know IQFeed has a reputation for being reliable, but after being disappointed with another comparable US equities data provider that seems to be generally viewed positively here, I'm beginning to suspect the concept of reliable may be very relative and would like to get more specific about reliability before I go switching to IQFeed. Perhaps the issue is that I just expect too much for the price point or that I'm more focused on historical data while everyone else is focused on live/feed data?
Anyway, here's a list of the sort of issues I've had in the last month with my current data provider that's making me consider switching, and hopefully you all can give me some perspective on whether this is just par for the course for a $100+/month data provider or not.
1.) Multiday period in which API requests for historical data would occasionally return incomplete results but in a way that gave you no easy way to tell the results were incomplete without repeating the request and noticing more data being returned. According to status everything was A-OK when partial results were returned.
2.) About a third of the time I manually check the Historical NBBO data against another source like IB, I quickly notice stretches of time missing NBBO updates that last for multiple seconds and sometimes minutes.
3.) Identical API calls occasionally (~once every 150 calls) returning different values. E.g. AMZN volume on date 2020-04-01 being reported as 4121875 and then seconds later an identical call reporting it as 282532153. So far this has lasted for at least a day, no word from vendor.
4.) Split adjustments for a few symbols I checked were backwards (e.g. multiplied by 2 when should have divided by 2, seemed systemic). For certain bar sizes the volume weighted price field would be adjusted and for others it wouldn't. Although it's been at least a month, the problem has not been fixed. Currently it seems selecting adjustments does nothing. For example the "adjusted" value for AAPL goes from the 600's to the 90's in one day after last split.
5.) volume weighted average bar prices that are larger than the bar's high or lower than the bar's low.
This is what I've noticed in just a little over a month of using the data provider. Is this normal for the price point? Does IQFeed have similar random problems?
Rant: honestly, for $100+/month is it too much to ask that the developers put in some basic unit tests so that they don't break what's already working and notice obvious problems in the data? Maybe I just don't recognize the inherent difficulty at doing all this at such a large data scale, but this seems crazy -- I don't see how people are comfortable trading on this data and really hope it's not the norm.
Anyway, here's a list of the sort of issues I've had in the last month with my current data provider that's making me consider switching, and hopefully you all can give me some perspective on whether this is just par for the course for a $100+/month data provider or not.
1.) Multiday period in which API requests for historical data would occasionally return incomplete results but in a way that gave you no easy way to tell the results were incomplete without repeating the request and noticing more data being returned. According to status everything was A-OK when partial results were returned.
2.) About a third of the time I manually check the Historical NBBO data against another source like IB, I quickly notice stretches of time missing NBBO updates that last for multiple seconds and sometimes minutes.
3.) Identical API calls occasionally (~once every 150 calls) returning different values. E.g. AMZN volume on date 2020-04-01 being reported as 4121875 and then seconds later an identical call reporting it as 282532153. So far this has lasted for at least a day, no word from vendor.
4.) Split adjustments for a few symbols I checked were backwards (e.g. multiplied by 2 when should have divided by 2, seemed systemic). For certain bar sizes the volume weighted price field would be adjusted and for others it wouldn't. Although it's been at least a month, the problem has not been fixed. Currently it seems selecting adjustments does nothing. For example the "adjusted" value for AAPL goes from the 600's to the 90's in one day after last split.
5.) volume weighted average bar prices that are larger than the bar's high or lower than the bar's low.
This is what I've noticed in just a little over a month of using the data provider. Is this normal for the price point? Does IQFeed have similar random problems?
Rant: honestly, for $100+/month is it too much to ask that the developers put in some basic unit tests so that they don't break what's already working and notice obvious problems in the data? Maybe I just don't recognize the inherent difficulty at doing all this at such a large data scale, but this seems crazy -- I don't see how people are comfortable trading on this data and really hope it's not the norm.
Last edited:
