How to make predict_in_sample()
derived from the first d
term of original data when differencing order (d
) isn't zero
#533
Labels
Describe the question you have
I think the the usage of
predict_in_sample()
inpmdarima
(python pkg) is same asfitted()
inforecast
(R pkg).But I find out that the output of
predict_in_sample()
inpmdarima
(python pkg) is different with output offitted()
inforecast
(R pkg) when difference order isn't zero.I use the following python codes to generate the output of
predict_in_sample()
in three different differencing order (d
) :I copy the output of above python codes, then paste to a R script to compare the difference between
predict_in_sample()
inpmdarima
andfitted()
inforecast
(R pkg), R script:output image of R script:
d
(difference order) term has huge difference, then they getting close afterd
(difference order) term.forecast
(R package) is correct., based on:d
term of original time-series(black line) and output offitted()
(red line)My questions are:
predict_in_sample()
inpmdarima
to get the same output withfitted()
inforecast
?start
, but it only shorten the length of output ofpredict_in_sample()
.Versions (if necessary)
No response
The text was updated successfully, but these errors were encountered: