| Interface | Description |
|---|---|
| ModelOptimizer |
Finds best model for given data set.
|
| TimeSeriesModel |
Basic interface for time series model
|
| Class | Description |
|---|---|
| AbstractExponentialSmoothing | |
| AbstractModelOptimizer | |
| ContinuousModel |
Some models throw exception if model has not been initialized
TimeSeriesModel.init(List) before calling
TimeSeriesModel.learn(DataPoint). |
| DoubleExponentialSmoothing |
Double exponential smoothing (Holt's linear trend model)
Works well when data exhibits increasing or decreasing trend pattern.
|
| DoubleExponentialSmoothing.Optimizer | |
| DoubleExponentialSmoothing.State | |
| SimpleExponentialSmoothing |
Simple exponential smoothing
Works well for stationary data when there is no trend in data.
|
| SimpleExponentialSmoothing.Optimizer | |
| SimpleExponentialSmoothing.State | |
| SimpleMovingAverage |
Simple moving average model, often used for trend estimating
|
| TimeSeriesModel.TimestampComparator | |
| TripleExponentialSmoothing |
Triple exponential smoothing model also known as Holt-Winters model.
|
| TripleExponentialSmoothing.Optimizer | |
| TripleExponentialSmoothing.State | |
| WeightedMovingAverage |
Weighted moving average model, variant of Moving average model.
|
| Enum | Description |
|---|---|
| Model |
Copyright © 2015–2016 Red Hat, Inc.. All rights reserved.