ridge2f: Ridge2 model

In Techtonique/ahead: Time Series Forecasting with uncertainty quantification

Ridge2 model

Description

Random Vector functional link network model with 2 regularization parameters

Usage

ridge2f(
  y,
  h = 5,
  level = 95,
  xreg = NULL,
  lags = 1,
  nb_hidden = 5,
  nodes_sim = c("sobol", "halton", "unif"),
  activ = c("relu", "sigmoid", "tanh", "leakyrelu", "elu", "linear"),
  a = 0.01,
  lambda_1 = 0.1,
  lambda_2 = 0.1,
  dropout = 0,
  type_forecast = c("recursive", "direct"),
  type_pi = c("gaussian", "bootstrap", "blockbootstrap", "movingblockbootstrap",
    "rvinecopula"),
  block_length = NULL,
  margins = c("gaussian", "empirical"),
  seed = 1,
  B = 100L,
  type_aggregation = c("mean", "median"),
  centers = NULL,
  type_clustering = c("kmeans", "hclust"),
  ym = NULL,
  cl = 1L,
  show_progress = TRUE,
  ...
)

Arguments

y	A univariate of multivariate time series of class ts (preferred) or a matrix
h	Forecasting horizon
level	Confidence level for prediction intervals
xreg	External regressors. A data.frame (preferred) or a matrix
lags	Number of lags
nb_hidden	Number of nodes in hidden layer
nodes_sim	Type of simulation for nodes in the hidden layer
activ	Activation function
a	Hyperparameter for activation function "leakyrelu", "elu"
lambda_1	Regularization parameter for original predictors
lambda_2	Regularization parameter for transformed predictors
dropout	dropout regularization parameter (dropping nodes in hidden layer)
type_forecast	Recursive or direct forecast
type_pi	Type of prediction interval currently "gaussian", "bootstrap", "blockbootstrap", "movingblockbootstrap", "rvinecopula" (with Gaussian and empirical margins for now)
block_length	Length of block for circular or moving block bootstrap
margins	Distribution of margins: "gaussian", "empirical" for type_pi == "rvinecopula"
seed	Reproducibility seed for random stuff
B	Number of bootstrap replications or number of simulations (yes, 'B' is unfortunate)
type_aggregation	Type of aggregation, ONLY for bootstrapping; either "mean" or "median"
centers	Number of clusters for type_clustering
type_clustering	"kmeans" (K-Means clustering) or "hclust" (Hierarchical clustering)
ym	Univariate time series (stats::ts) of yield to maturities with frequency = frequency(y) and start = tsp(y)[2] + 1 / frequency(y). Default is NULL.
cl	An integer; the number of clusters for parallel execution, for bootstrap
show_progress	A boolean; show progress bar for bootstrapping? Default is TRUE.
...	Additional parameters to be passed to kmeans or hclust

Value

An object of class "mtsforecast"; a list containing the following elements:

method	The name of the forecasting method as a character string
mean	Point forecasts for the time series
lower	Lower bound for prediction interval
upper	Upper bound for prediction interval
sims	Model simulations for bootstrapping (basic, or block)
x	The original time series
residuals	Residuals from the fitted model
coefficients	Regression coefficients for type_pi == 'gaussian' for now

Author(s)

T. Moudiki

References

Moudiki, T., Planchet, F., & Cousin, A. (2018). Multiple time series forecasting using quasi-randomized functional link neural networks. Risks, 6(1), 22.

Examples

require(fpp)

print(ahead::ridge2f(fpp::insurance)$mean)
print(ahead::ridge2f(fpp::usconsumption)$lower)

res <- ahead::ridge2f(fpp::insurance, h=10, lags=2)
par(mfrow=c(1, 2))
plot(res, "Quotes")
plot(res, "TV.advert")

# include a trend (just for the example)
xreg <- as.numeric(time(fpp::insurance))
res2 <- ahead::ridge2f(fpp::insurance, xreg=xreg,
h=10, lags=2)
par(mfrow=c(1, 2))
plot(res2, "Quotes")
plot(res2, "TV.advert")

# block bootstrap
xreg <- as.numeric(time(fpp::insurance))
res3 <- ahead::ridge2f(fpp::insurance, xreg=xreg,
                      h=10, lags=1L, type_pi = "bootstrap", B=10)
res5 <- ahead::ridge2f(fpp::insurance, xreg=xreg,
                      h=10, lags=1L, type_pi = "blockbootstrap", B=10,
                      block_length = 4)

print(res3$sims[[2]])
print(res5$sims[[2]])

par(mfrow=c(2, 2))
plot(res3, "Quotes")
plot(res3, "TV.advert")
plot(res5, "Quotes")
plot(res5, "TV.advert")


res4 <- ahead::ridge2f(fpp::usconsumption, h=20, lags=2L,
lambda_2=1)
par(mfrow=c(1, 2))
plot(res4, "income")
plot(res4, "consumption")


# moving block bootstrap
xreg <- as.numeric(time(fpp::insurance))
res6 <- ahead::ridge2f(fpp::insurance, xreg=xreg,
                      h=10, lags=1L,
                      type_pi = "movingblockbootstrap", B=10,
                      block_length = 4)

print(res6$sims[[2]])

par(mfrow=c(1, 2))
plot(res6, "Quotes")
plot(res6, "TV.advert")

Techtonique/ahead documentation built on Nov. 24, 2024, 10:33 a.m.