Returns predictions and weights calculated by sequential numeric optimization. The optimization is done stepwise, always calculating a one-step-ahead forecast.
batch(
y,
experts,
tau = 1:dim(experts)[2]/(dim(experts)[2] + 1),
affine = FALSE,
positive = FALSE,
intercept = FALSE,
debias = TRUE,
lead_time = 0,
initial_window = 30,
rolling_window = initial_window,
loss_function = "quantile",
loss_parameter = 1,
qw_crps = FALSE,
basis_knot_distance = 1/(dim(experts)[2] + 1),
basis_knot_distance_power = 1,
basis_deg = 1,
forget = 0,
soft_threshold = -Inf,
hard_threshold = -Inf,
fixed_share = 0,
p_smooth_lambda = -Inf,
p_smooth_knot_distance = basis_knot_distance,
p_smooth_knot_distance_power = basis_knot_distance_power,
p_smooth_deg = basis_deg,
p_smooth_ndiff = 1.5,
parametergrid_max_combinations = 100,
parametergrid = NULL,
forget_past_performance = 0,
allow_quantile_crossing = FALSE,
trace = TRUE
)
A numeric matrix of realizations. In probabilistic settings a matrix of dimension Tx1, in multivariate settings a TxP matrix. In the latter case, each slice of the expert's array gets evaluated using the corresponding column of the y matrix.
An array of predictions with dimension (Observations, Quantiles, Experts).
A numeric vector of probabilities.
Defines whether weights are summing to 1 or not. Defaults to FALSE.
Defines if a positivity constraint is applied to the weights. Defaults to FALSE.
Determines if an intercept is added, defaults to FALSE. If true, a new first expert is added, always predicting 1.
Defines whether the intercepts weight is constrained or not. If TRUE (the default), the intercept weight is unconstrained. Only affects the results if affine and or positive is set to TRUE. If FALSE, the intercept is treated as an expert.
offset for expert forecasts. Defaults to 0, which means that experts forecast t+1 at t. Setting this to h means experts predictions refer to t+1+h at time t. The weight updates delay accordingly.
Defines the size of the initial estimation window.
Defines the size of the rolling window. Defaults to the value of initial_window. Set it to the number of observations to receive an expanding window.
Either "quantile", "expectile" or "percentage".
Optional parameter scaling the power of the loss function.
Decides whether the sum of quantile scores (FALSE) or the quantile weighted CRPS (TRUE) should be minimized. Defaults to FALSE. Which corresponds to Berrisch & Ziel (2021)
determines the distance of the knots in the probability
basis. Defaults to 1 / (dim(experts)[2] + 1)
.
Parameter which defines the symmetry of the basis reducing the probability space. Defaults to 1, which corresponds to equidistant knots. Values less than 1 create more knots in the center, while values above 1 concentrate more knots in the tails.
Degree of the basis reducing the probability space. Defaults to 1.
Adds an exponential forgetting to the optimization. Past observations will get less influence on the optimization. Defaults to 0, which corresponds to no forgetting.
If specified, the following soft threshold will be applied to the weights: w = sgn(w)*max(abs(w)-t,0) where t is the soft_threshold parameter. Defaults to -inf, which means that no threshold will be applied. If all expert weights are thresholded to 0, a weight of 1 will be assigned to the expert with the highest weights prior to thresholding. Thus soft_threshold = 1 leads to the 'follow the leader' strategy if method is set to "ewa".
If specified, the following hard thresholding will be applied to the weights: w = w*(abs(w)>t) where t is the threshold_hard parameter. Defaults to -inf, which means that no threshold will be applied. If all expert weights are thresholded to 0, a weight of 1 will be assigned to the expert with the highest weight prior to thresholding. Thus hard_threshold = 1 leads to the 'follow the leader' strategy if method is set to "ewa".
Amount of fixed share to be added to the weights. Defaults to 0. 1 leads to uniform weights.
Penalization parameter used in the smoothing step. -Inf causes the smoothing step to be skipped (default).
determines the distance of the knots. Defaults to the value of basis_knot_distance. Corresponds to the grid steps when knot_distance_power = 1 (the default).
Parameter which defines the symmetry of the P-Spline basis. Takes the value of basis_knot_distance_power if unspecified.
Degree of the B-Spine basis functions. Defaults to the value of basis_deg.
Degree of the differencing operator in the smoothing equation. 1.5 (default) leads to shrinkage towards a constant. Can take values from 1 to 2. If a value in between is used, a weighted sum of the first and second differentiation matrix is calculated.
Integer specifying the maximum number of parameter combinations that should be considered. If the number of possible combinations exceeds this threshold, the maximum allowed number is randomly sampled. Defaults to 100.
User supplied grid of parameters. Can be used if not all combinations of the input vectors should be considered. Must be a matrix with 13 columns (online) or 12 columns batch with the following order: basis_knot_distance, basis_knot_distance_power, basis_deg, forget_regret, soft_threshold, hard_threshold, fixed_share, p_smooth_lambda, p_smooth_knot_distance, p_smooth_knot_distance_power, p_smooth_deg, p_smooth_ndiff, gamma.
Share of past performance not to be considered, resp. to be forgotten in every iteration of the algorithm when selecting the best parameter combination. Defaults to 0.
Shall quantile crossing be allowed? Defaults to false, which means that predictions are sorted in ascending order.
Print a progress bar to the console? Defaults to TRUE.
Returns weights and corresponding predictions. It is possible to impose a convexity constraint to the weights by setting affine and positive to TRUE.
if (FALSE) {
T <- 50 # Observations
N <- 2 # Experts
P <- 9 # Quantiles
prob_grid <- 1:P / (P + 1)
y <- rnorm(n = T) # Realized
experts <- array(dim = c(T, P, N)) # Predictions
for (t in 1:T) {
experts[t, , 1] <- qnorm(prob_grid, mean = -1, sd = 1)
experts[t, , 2] <- qnorm(prob_grid, mean = 3, sd = sqrt(4))
}
model <- batch(
y = matrix(y),
experts = experts,
p_smooth_lambda = 10
)
print(model)
plot(model)
autoplot(model)
}