Facebook Prophet saturation failling

Question

I´m using Prophet to predict sales and using the saturation parameter to not have negative values:

df_prof = df_prof.sample(1000) df_prof['cap'] = 6000 df_prof['floor'] = 10

m = Prophet(growth='logistic')

m.fit(df_prof)
future = m.make_future_dataframe(periods=365)
future['cap'] = 6000
future['floor'] = 10
forecast = m.predict(future)
fig1 = m.plot(forecast)

fig2 = m.plot_components(forecast)

But the predicted values have values outside both thresholds, lower and higher

Answer 1

it's because the forecasting method is not set to logistic.