TypeError：ufunc'add'不包含带有ARIMA模型签名的循环

Question

I trying to do ARIMA model analysis on a series extracted from the following pandas DataFrame

CpcGDP.tail()

               65
2012-01-01  2593.23
2013-01-01  2591.06
2014-01-01  2608.38
2015-01-01  2665.35
2016-01-01  2724.4

I first convert objects

CpcGDP.convert_objects(convert_numeric= True)

Then get the series and convert to float

Cdx = CpcGDP.columns[0]
S = CpcGDP.loc[:, Cdx]
S.astype(float)

Then do the analysis

 mod = sm.tsa.ARIMA(S.astype(float), order= (0,2,1))
 result = mod.fit()

But I getthe following error

 TypeError: ufunc 'add' did not contain a loop with signature matching types dtype('<U21') dtype('<U21') dtype('<U21')

Can you help me solve this error, please?

Answer 1

Had the same error doing the same stuff. For me the solution was using pandas Series instead of Dataframe