直接预测

使用周期性采样的时间序列数据执行直接预测

使用 directforecaster 函数基于周期性采样的时间序列数据训练 DirectForecaster 模型。Statistics and Machine Learning Toolbox™ 函数创建一个使用直接策略的多步预测模型，其中为预测期的每一步训练一个单独的回归模型。

在创建 DirectForecaster 模型对象后，您可以通过使用 loss 和 predict 对象函数来查看该模型在观测的测试数据上的表现。然后，您可以使用该模型通过 forecast 对象函数预测可用数据之外的时间步上的数据。

函数

`compact`	Reduce size of direct forecasting model (自 R2023b 起)
`crossval`	Cross-validate direct forecasting model (自 R2023b 起)
`loss`	Loss at each horizon step (自 R2023b 起)
`predict`	Predict response at time steps in observed test data (自 R2023b 起)
`forecast`	Forecast response at time steps beyond available data (自 R2023b 起)
`preparedPredictors`	Obtain prepared data used for training or testing in direct forecasting (自 R2023b 起)
`cvloss`	Loss for partitioned data at each horizon step (自 R2023b 起)
`cvpredict`	Predict response using cross-validated direct forecasting model (自 R2023b 起)
`tspartition`	Partition time series data for cross-validation (自 R2022b 起)

`DirectForecaster`	Fit direct forecasting model (自 R2023b 起)
`CompactDirectForecaster`	Compact direct forecasting model (自 R2023b 起)
`PartitionedDirectForecaster`	Cross-validated direct forecasting model (自 R2023b 起)

Perform Time Series Direct Forecasting with directforecaster
Perform time series direct forecasting with the directforecaster function.
Manually Perform Time Series Forecasting Using Ensembles of Boosted Regression Trees
Manually perform single-step and multiple-step time series forecasting with ensembles of boosted regression trees.