pca.elm_forecast {ICompELM}

Forecasting from PCA based ELM model

Description

Forecasts are generated recursively from a trained Extreme Learning Machine built using Principal Component Analysis.

Usage

pca.elm_forecast(pca.elm_model, h = 1)

Arguments

pca.elm_model

A trained PCA based ELM model.

h

Number of periods for forecasting. Defaults to one-step ahead forecast.

Value

Vector of point forecasts.

See Also

pca.elm_train() for training an ICA based ELM model.

Examples

train_set <- head(price, 12*12)
test_set <- tail(price, 12)
pca.model <- pca.elm_train(train_data = train_set, lags = 12)
y_hat <- pca.elm_forecast(pca.elm_model = pca.model, h = length(test_set))
# Evaluation of the forecasts
if(require("forecast")) forecast::accuracy(y_hat, test_set)