I am aiming to execute a regression analysis on the diffusion of tweets compared to hashtags that are incorporated in tweets.
When executing this code
#Changing X and y to arrays using Numpy#
x = np.array(hashtag_df['tagscore'])
y = np.array(hashtag_df['diffusion_count'])
#Adding a constant to x to execute the analysis
x = sm.add_constant(x)
#Create and fit model for the analysis
modelsm = sm.OLS(y, X)
smresults = modelsm.fit()
#Print a summary of the model results
print (smresults.summary())
the following error pops up
TypeError: ufunc 'isfinite' not supported for the input types, and the inputs could not be safely coerced to any supported types according to the casting rule ''safe''
From other questions I noticed this was a dtype error and that I had to change my dtype to float. However both tagscore as well as diffusion_count are already floats
hashtag_df.dtypes
id int64
taglist object
diffusion_count float64
tagscore float64
dtype: object
Where does this error arise? and how can it be resolved?