Mean absolute error
Obviously, before you use the model to predict, you want to know how accurate your predictions are. The mean absolute error (MAE) is a good statistic for this. It is the mean difference between your predictions and the true values.
In this exercise you will calculate the MAE for an ARMA(1,1) model fit to the earthquakes time series
numpy
has been imported into your environment as np
and the earthquakes time series is available for you as earthquake
.
This exercise is part of the course
ARIMA Models in Python
Exercise instructions
- Use
np
functions to calculate the Mean Absolute Error (MAE) of the.resid
attribute of theresults
object. - Print the MAE.
- Use the DataFrame's
.plot()
method with no arguments to plot theearthquake
time series.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Fit model
model = ARIMA(earthquake, order=(1,0,1))
results = model.fit()
# Calculate the mean absolute error from residuals
mae = ____
# Print mean absolute error
print(____)
# Make plot of time series for comparison
____
plt.show()