'model weekly seasonality with fourier terms in python

I've hourly data and I want to model weekly seasonality with fourier terms. I'd like to implement it in pyhton.

I was trying in this way but I'm not sure about it.

exog = pd.DataFrame({'date': df.index})
exog = exog.set_index(df.index)

i = 4  
for k in range(1, i):
    exog['week_sin'+str(k)] = np.sin(2 *k* np.pi * exog.index.hour/168)  
    exog['week_cos'+str(k)] = np.cos(2 *k* np.pi * exog.index.hour/168)

python time-series

Sources

This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.

Source: Stack Overflow

Solution	Source

'model weekly seasonality with fourier terms in python

Sources

Related Questions