dbnR (version 0.7.8)

forecast_ts: Performs forecasting with the GDBN over a dataset

Description

Given a dbn.fit object, the size of the net and a folded dataset, performs a forecast over the initial evidence taken from the dataset.

Usage

forecast_ts(
  dt,
  fit,
  size = NULL,
  obj_vars,
  ini = 1,
  len = dim(dt)[1] - ini,
  rep = 1,
  num_p = 50,
  print_res = TRUE,
  plot_res = TRUE,
  mode = "exact",
  prov_ev = NULL
)

Value

a list with the original time series values and the results of the forecast

Arguments

dt

data.table object with the TS data

fit

dbn.fit object

size

number of time slices of the net. Deprecated, will be removed in the future

obj_vars

variables to be predicted

ini

starting point in the dataset to forecast.

len

length of the forecast

rep

number of times to repeat the approximate forecasting

num_p

number of particles in the approximate forecasting

print_res

if TRUE prints the mae and sd metrics of the forecast

plot_res

if TRUE plots the results of the forecast

mode

"exact" for exact inference, "approx" for approximate

prov_ev

variables to be provided as evidence in each forecasting step

Examples

size = 3
data(motor)
dt_train <- motor[200:900]
dt_val <- motor[901:1000]
obj <- c("pm_t_0")
net <- learn_dbn_struc(dt_train, size)
f_dt_train <- fold_dt(dt_train, size)
f_dt_val <- fold_dt(dt_val, size)
fit <- fit_dbn_params(net, f_dt_train, method = "mle-g")
res <- suppressWarnings(forecast_ts(f_dt_val, fit, 
        obj_vars = obj, len = 10, print_res = FALSE, plot_res = FALSE))