If you look at the documentation for STL(): https://www.statsmodels.org/stable/examples/notebooks/generated/stl_decomposition.htmlSTL()
You, you can see that there is a default of 7 for the seasonal period, if you do not have a frequency (which is what I think you have here).
The decomposition requires 1 input, the data series. If the data series does not have a frequency, then you must also specify period. The default value for seasonal is 7, and so should also be changed in most applications.
Hopefully, this will help!
I have also faced this issue when using stats model:
import statsmodels.api as sm
decomposition = sm.tsa.seasonal_decompose(df, model='additive')
fig = decomposition.plot()
I get the following error:
ValueError: You must specify importa statsmodels.apiperiod asor sm
x must be a pandas decompositionobject =with sm.tsa.seasonal_decompose(df,a model='additive')
DatetimeIndex with a freq fignot =set decomposition.plot()to None
I get the following error:
ValueError: You must specify a period or x must be a pandas object with a DatetimeIndex with a freq not set to None
However, when I set a period of 2, it works fine.
import statsmodels.api as sm
decomposition = sm.tsa.seasonal_decompose(df, period = 2,
model='additive')
fig = decomposition.plot()
import statsmodels.api as sm
decomposition = sm.tsa.seasonal_decompose(df, period = 2,
model='additive')
fig = decomposition.plot()
The problem is, I have only yearly data, without monthly information, and no clear period to work with. Therefore, it is arbitrary (perhaps incorrect) to set a random period for my data set. If you do have a period (some seasonal time span) that you know of in your data, you can use this,this; otherwise, I am not sure how to help.
See further the issue discussed here: decompose() for time series: ValueError: You must specify a period or x must be a pandas object with a DatetimeIndex with a freq not set to Nonehere.
