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
Now I am using the ARIMA model in the sparkts library. However, when I was reviewing the code on github, I found some errors about function autoFit() in ARIMA.scala
autoFit(ts, maxP, maxD, maxQ)
* @param ts time series to which to automatically fit an ARIMA model
* @param maxP limit for the AR order
* @param maxD limit for differencing order
* @param maxQ limit for the MA order
* @return an ARIMAModel
As you can see above, you can see the role of each parameter.
So, when I did autoFit(ts,0,0,0) I thought that results were p = 0, q = 0 because maxP = 0 and maxQ = 0.
But result was maxP = 2, maxQ = 2.
I don't know why these results came out.
In the picture I attached, the maxD, maxP, maxQ values are specified in the code to limit the value. In this code, although the max value is specified, the result is higher than the max value.I can not be sure if I do not understand or if there is an error in this code.
I'll wait for your reply.
Thank you.
The text was updated successfully, but these errors were encountered:
Hi there,
Now I am using the ARIMA model in the sparkts library. However, when I was reviewing the code on github, I found some errors about function autoFit() in ARIMA.scala
autoFit(ts, maxP, maxD, maxQ)
As you can see above, you can see the role of each parameter.
So, when I did autoFit(ts,0,0,0) I thought that results were p = 0, q = 0 because maxP = 0 and maxQ = 0.
But result was maxP = 2, maxQ = 2.
I don't know why these results came out.
In the picture I attached, the maxD, maxP, maxQ values are specified in the code to limit the value. In this code, although the max value is specified, the result is higher than the max value.I can not be sure if I do not understand or if there is an error in this code.
I'll wait for your reply.
Thank you.
The text was updated successfully, but these errors were encountered: