You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Quandl skips/rounds the decimal points on my API calls.
For example futures = Quandl("CFTC/IC6_F_ALL_CR")
Returns rounded 100, 99, or 98 where all the information after floating-point is lost. And yes, it's not "hiden", but the precision after floating-point is lost, this is obvious when one plots the data or multiplies the series by 100.
The same problem applies for other series as well (for example: "OPEC/ORB").
I do have European number settings in my Ubuntu 20.04 compter (e.g thousands are separated by space and floating points by comma on most occasions), but I have not used this setting for R and this has no effect on my other R code/other libraries. E.g the floating points are still separated by "." in R and the other code works without problems.
Quandl skips/rounds the decimal points on my API calls.
For example
futures = Quandl("CFTC/IC6_F_ALL_CR")
Returns rounded 100, 99, or 98 where all the information after floating-point is lost. And yes, it's not "hiden", but the precision after floating-point is lost, this is obvious when one plots the data or multiplies the series by 100.
The same problem applies for other series as well (for example: "OPEC/ORB").
I do have European number settings in my Ubuntu 20.04 compter (e.g thousands are separated by space and floating points by comma on most occasions), but I have not used this setting for R and this has no effect on my other R code/other libraries. E.g the floating points are still separated by "." in R and the other code works without problems.
My current hypothesis is that the problem is related to jsonlite library (see: https://stackoverflow.com/questions/36038349/significant-digits-when-converting-json-using-fromjson-in-jsonlite). I will fork the Quandl library and test, whether that's the culprit.
The text was updated successfully, but these errors were encountered: