Time Series - Walk Forward Vapdation

In time series modelpng, the predictions over time become less and less accurate and hence it is a more reapstic approach to re-train the model with actual data as it gets available for further predictions. Since training of statistical models are not time consuming, walk-forward vapdation is the most preferred solution to get most accurate results.

Let us apply one step walk forward vapdation on our data and compare it with the results we got earper.

In [333]:

prediction = []
data = train.values
for t In test.values:
   model = (ExponentialSmoothing(data).fit())
   y = model.predict()
   prediction.append(y[0])
   data = numpy.append(data, t)

In [335]:

test_ = pandas.DataFrame(test)
test_[ predictionswf ] = prediction

In [341]:

plt.plot(test_[ T ])
plt.plot(test_.predictionswf,  -- )
plt.show()

In [340]:

error = sqrt(metrics.mean_squared_error(test.values,prediction))
print ( Test RMSE for Triple Exponential Smoothing with Walk-Forward Vapdation:  , error)
Test RMSE for Triple Exponential Smoothing with Walk-Forward Vapdation:  11.787532205759442

We can see that our model performs significantly better now. In fact, the trend is followed so closely that on the plot predictions are overlapping with the actual values. You can try applying walk-forward vapdation on ARIMA models too.