Incomplete Time Series Prediction Using Max-Margin Classification of Data with Absent Features

Figure 2

The implementation process of our model.