Hello everybody,
does someone uses this data and can tell us something about the quality of the data.
For me it looks interesting to get bundle including historical intraday data for 15+ years.
https://firstratedata.com/b/22/stock-complete-historical-intraday
https://firstratedata.com/bundle/all
What is your opinion about the data, the provider and the pricing?
does someone uses this data and can tell us something about the quality of the data.
For me it looks interesting to get bundle including historical intraday data for 15+ years.
https://firstratedata.com/b/22/stock-complete-historical-intraday
https://firstratedata.com/bundle/all
What is your opinion about the data, the provider and the pricing?
Rename
I would like to update this topic again before it disappears into history.
On the main page https://firstratedata.com/ WealthLab is even mentioned I noticed.
After a short research, my impression is that it is a reputable provider and the data quality seems to be decent. The prices for the one-time payment for the historical data (with one month update bonus) seems to be fair.
The update subscription has to be paid separately. This is good, in my opinion, because most already obtain current data from the broker of their choice. Having the ability to keep the data up to date is also good, of course. It is difficult for me to judge whether the monthly price is okay.
On the main page https://firstratedata.com/ WealthLab is even mentioned I noticed.
After a short research, my impression is that it is a reputable provider and the data quality seems to be decent. The prices for the one-time payment for the historical data (with one month update bonus) seems to be fair.
The update subscription has to be paid separately. This is good, in my opinion, because most already obtain current data from the broker of their choice. Having the ability to keep the data up to date is also good, of course. It is difficult for me to judge whether the monthly price is okay.
@MIH - I have used FirstRate data for specific historical intraday tests.
Data quality seems ok - although I have only tested a few symbols
FR include 1m, 5m, 30m and 1hr intraday data. I have found parsing 1m into WL8 is sufficient for me, since WL8 will let you pick any other period from your original data. However loading 10+ years of 1m data (~1.6mm bars) can be slow and draw hard on memory.
If you are planning on large dataset backtests with intraday data, then you'll be better off loading the most appropriate data scale to save loading time.
Parsing FR data into WL8 is slow - given the large amount of ascii data. It is however a one-time exercise. Good thing Eugene and Glitch implemented an Ascii cache - this makes accessing the data faster after the 2nd time
Am considering a memory upgrade to my pc (currently on 16gb) to better cope with 1min x 10yr optimizations :)
Have not taken up the renewal service for the reasons you mentioned...
Eventually we should be able to merge current intraday data service providers with FR historical data to update the dataset. Assume an ascii export, then append in windows will do the job. Would welcome any suggestions on how to do better
Data quality seems ok - although I have only tested a few symbols
FR include 1m, 5m, 30m and 1hr intraday data. I have found parsing 1m into WL8 is sufficient for me, since WL8 will let you pick any other period from your original data. However loading 10+ years of 1m data (~1.6mm bars) can be slow and draw hard on memory.
If you are planning on large dataset backtests with intraday data, then you'll be better off loading the most appropriate data scale to save loading time.
Parsing FR data into WL8 is slow - given the large amount of ascii data. It is however a one-time exercise. Good thing Eugene and Glitch implemented an Ascii cache - this makes accessing the data faster after the 2nd time
Am considering a memory upgrade to my pc (currently on 16gb) to better cope with 1min x 10yr optimizations :)
Have not taken up the renewal service for the reasons you mentioned...
Eventually we should be able to merge current intraday data service providers with FR historical data to update the dataset. Assume an ascii export, then append in windows will do the job. Would welcome any suggestions on how to do better
Thanks for your feedback.
Your comment about how effectively the data can be used in WL is valuable. However, it looks like the amount of data can be used effectively and WL can handle it.
I am even of the opinion that for the frequency of one minute not necessarily 15 years must be used. The high year numbers are mainly useful for the higher frequencies. Of course, it is great if 10 years also run smoothly with the 1 minute frequency.
The longer I think about the update subscription the better I find it, even if data from another source (for me IB) is available. The reason is that the completeness is guaranteed. With a bundle of 7000+ symbols the time restriction but also the completeness of the symbols is not guaranteed by the broker. Updating all symbols daily for all time units does not work through the broker, according to my experience so far.
The price of $34.95 per month adds up, but when I consider what else I spend 35 bucks a month on, it all comes into proportion. (Internet connection, Mobile phone, Video on demand and so on ...) In the end, complete data is available for all symbols, which can be downloaded quickly. (Instead of being retrieved via individual API requests and broker limitations).
Your comment about how effectively the data can be used in WL is valuable. However, it looks like the amount of data can be used effectively and WL can handle it.
I am even of the opinion that for the frequency of one minute not necessarily 15 years must be used. The high year numbers are mainly useful for the higher frequencies. Of course, it is great if 10 years also run smoothly with the 1 minute frequency.
The longer I think about the update subscription the better I find it, even if data from another source (for me IB) is available. The reason is that the completeness is guaranteed. With a bundle of 7000+ symbols the time restriction but also the completeness of the symbols is not guaranteed by the broker. Updating all symbols daily for all time units does not work through the broker, according to my experience so far.
The price of $34.95 per month adds up, but when I consider what else I spend 35 bucks a month on, it all comes into proportion. (Internet connection, Mobile phone, Video on demand and so on ...) In the end, complete data is available for all symbols, which can be downloaded quickly. (Instead of being retrieved via individual API requests and broker limitations).
Your Response
Post
Edit Post
Login is required