Which data vendors provide historical data in good quality for US stocks?

I have read that Norgate are the best, but it is too expensive for me at the moment. So, I would accept, when it is not survivor-ship bias free. But stock splits should be calculated right, and the data should be correct mostly.

Which data vendor do provide historical data in good quality for US stocks (and European stocks perhaps) based on your own experience?

If you need daily bars, Norgate is probably the best. They always record the splits accurately before IBKR does. I had trouble with Yahoo's free data and with IBKR's data when stocks split.

1 Like

Yes, Norgate is probably the best, but to expensive. And about yahoo... I think, we don't need to talk :wink: Is there no solution in the middle?

@EveryDayBetter, in general, you get what you pay for.

In other forums/communities I read dfferent opinions about the quality/correctness of some "cheap" providers (I mean the ones that cost under 20$ per month).

Some traders compared their data to more established or institutional sources - especially going back multiple years - and reported incorrect dividends, penny rounding where stocks traded in 1/8s, missed application of dividends to divs adjusted prices, and/or plain wrong unadjusted raw prices for deep history.
(In some cases, I also have personally seen stock prices - for some isolated dates - with the decimal separator in the wrong place).

Anyway, at least in the opinions I read, USA stock quotes seem more reliable than the European ones.

If your trading strategies are heavily dependent on accurate historical prices, I suggest trying the different services (compare also to Norgate) to see if the data variations are small enough to still give you a potential edge.

2 Likes

Thank you for your answer. Do you have read something about Tiingo?

@EveryDayBetter, as I said, US stock data seems more reliable than the rest of the world. In particular I saw a trader who compared a specific stock (Citigroup), using FactSet as a reference against Tiingo and FMP, and he wrote: "Of the two Tiingo is better."

In any case, you should keep in mind that any comparison was made at a specific time, and it is possible that the data providers subsequently may have corrected their data (such as missed splits, dividends, etc.).

For example, in my experience, EODHistoricalData is quite fast in correcting wrong data if it is reported to them, but it is still a significant drawback (especially if you do not notice soon the inconsistency of the data provided).

In my view, any data provider should have a robust mechanism for detecting anomalies and the ability to check the consistency of their data sources constantly. Personally, I'm not convinced that many of them (cheap or free) can guarantee it.*

Anyway, the importance of high accuracy of past data is a subject that can be controversial.

While it would be interesting to hear the experiences of other users who use these providers specifically with Amibroker, I still think it is better to do your own evaluation and draw your own conclusions.

* Any data provider reading this personal point of view is cordially invited to contradict me by replying in this thread specifying how their data is continually verified and corrected.

4 Likes

Thank you very much for sharing your thoughts with me.

Norgate....the best I've ever seen. But only EOD. Norgate also provides historical index compositions. Most traders underestimate the "Rate of Survivorship" effect in historical back tests over long periods of time. Norgate offers the highest quality at what I think is a very fair price.

3 Likes

For any users of Tiingo EOD historical data, have you been pleased with their service and the quality of the data? New to Amibroker and considering the paid service. Download through AmiQuote seems very fast. Appears better than Yahoo Finance or Stooq. Would appreciate any user experiences/input.

Thanks for noticing - my background is institutional quant trading (both at Citi and a hedge fund). I built the company (Tiingo) with an emphasis on rigorous data cleaning methodologies that automatically detect anomalies, and in case of human intervention, have a strict audit trail so a human has to write why a correction was made and when it was made. Our primary clients are enterprises, and we use those funds to give our data at cost, often a loss, to individuals as it's part of the purpose of giving back. I think sometimes people attribute cost to quality and my goal is to break that association. Thanks for sharing your friend's feedback.

5 Likes

Rishi thanks for giving us background info on the work you do. It is appreciated.

Hi Tomasz,

If its not too much to ask, can you please update this website page with Tiingo? I had no idea about the data service as it wasn't listed here: How to get quotes from various markets

Thanks in advance.