Rfssa (version 3.2.0)

funts: Functional Time Series (funts) Class

Description

The `funts` class is designed to encapsulate functional time series objects, including both univariate (FTS) and multivariate (MFTS) forms. It provides a versatile framework for creating and manipulating `funts` objects, accommodating various basis systems and dimensions.

Usage

funts(
  X,
  basisobj,
  argval = NULL,
  method = "data",
  start = 1,
  end = NULL,
  vnames = NULL,
  dnames = NULL,
  tname = NULL
)

Value

An instance of the `funts` class containing functional time series data.

An instance of the `funts` class containing functional time series data.

Arguments

X

A matrix, three-dimensional array, or a list of matrix or array objects. When `method="data"`, it represents the observed curve values at discrete sampling points or argument values. When `method="coefs"`, `X` specifies the coefficients corresponding to the basis system defined in 'basisobj'. If `X` is a list, it defines a multivariate FTS, with each element being a matrix or three-dimensional array object. In matrix objects, rows correspond to argument values, and columns correspond to the length of the FTS. In three-dimensional array objects, the first and second dimensions correspond to argument values, and the third dimension to the length of the FTS.

basisobj

An object of class `basisfd`, a matrix of empirical basis, or a list of `basisfd` or empirical basis objects. For empirical basis, rows correspond to basis functions, and columns correspond to grid points.

argval

A vector list of length `p` which is a set of argument values corresponding to the observations in X. Each entry in this list should either be a numeric value or a list of numeric elements, depending on the dimension of the domain the variable is observed over. It can even vary from one variable to another, If it be NULL, the default value for argval are the integers 1 to n, where n is the size of the first dimension in argument X.?

method

Determines the type of the `X` matrix: "coefs" or "data."

start

The time of the first observation. It can be a single positive integer or an object of classes `Date`, `POSIXct`, or `POSIXt`, specifying a natural time unit.

end

The time of the last observation, specified in the same way as `start`.

vnames

a vector of strings specifies the variable names

dnames

list of vector of strings specifies the variable domain names

tname

a string specifies the time index name

See Also

fssa

Examples

# 1D FTS example: Callcenter dataset
require(fda)
Call_data <- loadCallcenterData()
D <- matrix(sqrt(Call_data$calls), nrow = 240)
bs1 <- create.bspline.basis(c(0, 23), 22)
Y <- funts(X = D, basisobj = bs1, start = as.Date("1999-1-1"),
           vnames = "Sqrt of Call Numbers",
           dnames = "Time (6 minutes aggregated)",
           tname = "Date" )

plot(Y, lwd = 2, npts = 200, col = "deepskyblue4",
  main = "Call Center Data"
)

# _______2D Multivariate Example: Montana dataset
# Temperature curves and smoothed images of vegetation
montana_data <- loadMontanaData()
Temp <- montana_data$Temp
NDVI <- montana_data$NDVI
Montana_Data <- list(Temp / sd(Temp), NDVI)
bs1 <- create.bspline.basis(c(0, 23), 11)
bs2 <- create.bspline.basis(c(1, 33), 13)
bs2d <- list(bs2, bs2)
bsmv <- list(bs1, bs2d)
Y <- funts(X = Montana_Data, basisobj = bsmv,
           start = as.Date("2008-01-01"),
           end = as.Date("2013-09-30"),
           vnames = c("Normalized Temperature (\u00B0C)" , "NDVI"),
           dnames = list("Time", c("Latitude", "Longitude")),
           tname = "Date"
)

plot(Y,  main = "Montana dataset")