Automatic Forecasting Models
Automatic forecasting tools optimize the hyperparameters of a given
model class and select the best-performing model for a validation set.
The optimization methods include grid search, random search, and
Bayesian optimization.
Optimization Objectives
NeuralForecast is a highly modular framework capable of augmenting a
wide variety of robust neural network architectures with different point
or probability outputs as defined by their optimization objectives.
MLP-Based Model Family
The MLP-based family operates like a classic autoencoder. Its initial
layers encode raw autoregressive window into a representation, and the
decoder produces the desired output based on the horizon, probability
output, or point objective. Recent architectures include modifications
like residual learning techniques and task-specific changes.
Model | Point Forecast | Probabilistic Forecast | Insample fitted values | Probabilistic fitted values |
---|
MLP | β
| β
| β
| β
|
NBEATS | β
| β
| β
| β
|
NBEATSx | β
| β
| β
| β
|
NHITS | β
| β
| β
| β
|
RNN-Based Model Family
The RNN-based family attempts to leverage the dataβs temporal structure
while reducing MLPs over parametrization. Recurrent networks are dynamic
and can handle sequences of varying lengths through a mechanism for
updating internal states that considers the entire sequence history.
Modern state modifications help diminish vanishing and exploding
gradients.
Model | Point Forecast | Probabilistic Forecast | Insample fitted values | Probabilistic fitted values |
---|
RNN | β
| β
| β
| β
|
GRU | β
| β
| β
| β
|
LSTM | β
| β
| β
| β
|
TCN | β
| β
| β
| β
|
DeepAR | β
| β
| β
| β
|
DilatedRNN | β
| β
| β
| β
|
Transformer architectures are an alternative to recurrent networks.
These networks build on the self-attention mechanism that directly
allows modeling the relationship between different sequence parts
without sequential processing. Attention makes Transformers more
parallelizable than RNNs.
CNN-Based Model Family
Convolutional Neural Networks (CNNs), originally celebrated for their
accomplishments in image processing and computer vision, have also
revealed substantial prowess in time series forecasting. Navigating
through temporal data, CNNs utilize their convolutional layers to
automatically and adaptively learn temporal patterns from the input
data, offering an approach to uncovering subtle, underlying patterns
embedded within a series of values.
Model | Point Forecast | Probabilistic Forecast | Insample fitted values | Probabilistic fitted values |
---|
TimesNet | β
| β
| β
| β
|