Similarity: Similarity

In kdwang1808/Dejavu: Deja vu: forecasting with similarity

Description

Similarity

Usage

Similarity(
  y,
  fh,
  LoadData,
  path,
  nts = 100,
  Preprocessing = 1,
  Dist_type = 3,
  gamma = 0.05,
  scaling = c("sdiff", "snaive"),
  clustering = 0,
  FullOutput = FALSE
)

Arguments

y	"vector" the observation series
fh	number of forecasting horizon
LoadData
nts	number of the most similar series
Preprocessing	remove seasonality, smooth the series and scale the target and the reference series to the same magnitude
Dist_type	number of forecasting benchmarks
gamma	calibrated significance level quantiles
scaling	scaling method: using different scaling or not
clustering	number of sample categories in kmeans
FullOutput

Value

A list containing calculation of similarity

kdwang1808/Dejavu documentation built on March 23, 2020, 5:14 a.m.