Predicting ETH price using python

2 janv. 2022

Predicting ETH price for the next 30 days by using Facebook prophet in a simple manner.
Black dots represent reality, red lines are the changepoints in the trend, and the blue line is the prediction.



Here is the code:


###################
import pandas as pd
from fbprophet import Prophet

df.rename(columns={'Date': 'ds', 'Price (Close)': 'y'}, inplace=True)
df['ds'] = pd.to_datetime(df['ds'], errors='coerce', utc=True )
df['ds'] = df['ds'].dt.strftime('%Y-%m-%d %H:%M')
df.rename(columns={'Date': 'ds', 'Value': 'y'}, inplace=True)

m = Prophet(seasonality_mode='multiplicative')
m.add_seasonality(name='monthly', period=30.5, fourier_order=5)

m.fit(df)
future = m.make_future_dataframe(periods=30*96, freq='15min')
forecast = m.predict(future)
m.plot(forecast)
m.plot_components(forecast)
from fbprophet.plot import add_changepoints_to_plot
fig = m.plot(forecast)
add_changepoints_to_plot(fig.gca(), m, forecast)
plt.show(trend=True)


#################
Please share your feedback, especially if you see mistakes in the process.
And remember:
There is always a possibility for error.