Nothing
R/InferenceDataObject.R
In fdaPDE: Physics-Informed Spatial and Functional Data Analysis
Defines functions
inferenceDataObjectBuilder
Documented in
inferenceDataObjectBuilder
#' @useDynLib fdaPDE
#' @importFrom methods new
#' @importFrom stats pchisq pnorm qchisq qnorm
NULL
#' Class for inference data
#'
#'@slot test A vector of integers taking value 0, 1 or 2; if 0 no test is performed, if 1 one-at-the-time tests are performed, if 2 a simultaneous test is performed.
#'@slot interval A vector of integers taking value 0, 1, 2 or 3; if 0 no confidence interval is computed, if 1 one-at-the-time confidence intervals are computed,
#'if 2 simultaneous confidence intervals are computed, if 3 Bonferroni confidence intervals are computed.
#'@slot type A vector of integers taking value 1, 2, 3, 4 or 5 corresponding to Wald, Speckman, Eigen-Sign-Flip, Enhanced-Eigen-Sign-Flip or Sign-Flip inferential approach.
#'@slot component A vector of integers taking value 1, 2 or 3, indicating whether the inferential analysis should be carried out respectively for the parametric, nonparametric or both the components.
#'@slot exact An integer taking value 1 or 2. If 1 an exact computation of the test statistics will be performed,
#'whereas if 2 an approximated computation will be carried out (not implemented in this version).
#'@slot dim Dimension of the problem, it is equal to 2 in the 1.5D and 2D cases and equal to 3 in the 2.5D and 3D cases.
#'@slot n_cov Number of covariates taken into account in the linear part of the regression problem.
#'@slot locations A matrix of numeric coefficients with columns of dimension \code{dim}. When nonparametric inference is requested it represents the set of spatial locations for which the inferential analysis should be performed.
#' The default values is a one-dimensional matrix of value 1 indicating that all the observed location points should be considered.
#' In the sign-flip and eigen-sign-flip implementations only observed points are allowed.
#'@slot locations_indices A vector of indices indicating which spatial points have to be considered among the observed ones for nonparametric inference. If a vector of location indices is provided
#'then the slot 'location' is discarded.
#'@slot locations_are_nodes An integer taking value 1 or 2; in the first case it indicates that the selected locations to perform inference on f are all coinciding with the nodes; otherwise it takes value 2;
#'@slot coeff A matrix of numeric coefficients with columns of dimension \code{n_cov} and each row represents a linear combination of the linear parameters to be tested and/or to be
#' estimated via confidence interval.
#'@slot beta0 Vector of null hypothesis values for the linear parameters of the model. Used only if \code{test} is not 0 and \code{component} is not 2.
#'@slot f0 Function representing the expression of the nonparametric component f under the null hypothesis. Used only if \code{component} is not 1.
#'@slot f0_eval Vector of f0 evaluations at the choosen test locations. It will be eventually set later in checkInferenceParameters, if nonparametric inference is required.
#'@slot f_var An integer taking value 1 or 2. If 1 local variance estimates for the nonlinear part of the model will be computed,
#'whereas if 2 they will not.
#'@slot quantile Vector of quantiles needed for confidence intervals, used only if interval is not 0.
#'@slot alpha 1 minus confidence level vector of sign-flipping approaches confidence intervals. Used only if interval is not 0.
#'@slot n_flip An integer representing the number of sign-flips in the case of sign-flipping approaches.
#'@slot tol_fspai A real number greater than 0 specifying the tolerance for FSPAI algorithm, in case of non-exact inference (not implemented in this version).
#'@slot definition An integer taking value 0 or 1. If set to 1, the class will be considered as created by the function \code{\link{inferenceDataObjectBuilder}},
#'leading to avoid some of the checks that are performed on inference data within smoothing functions.
#'
#'@description
#' A class that contains all possible information for inference over linear parameters and/or nonparametric field in spatial regression with
#' differential regularization problem. This object can be used as parameter in smoothing function of the fdaPDE library \code{\link{smooth.FEM}}.
#'
#'@details
#'At least one between test and interval must be nonzero. \code{n_cov}, \code{coeff} and \code{beta0}, if provided, need to be coherent.
#'\code{dim} and \code{locations}, if provided, need to be coherent.
#'The usage of \code{\link{inferenceDataObjectBuilder}} is recommended for the construction of an object of this class.
inferenceDataObject<-setClass("inferenceDataObject", slots = list(test = "integer",
interval = "integer",
type = "integer",
component = "integer",
exact = "integer",
dim = "integer",
n_cov = "integer",
locations = "matrix",
locations_indices = "integer",
locations_are_nodes = "integer",
coeff = "matrix",
beta0 = "numeric",
f0 = "function",
f0_eval = "numeric",
f_var = "integer",
quantile = "numeric",
alpha = "numeric",
n_flip = "integer",
tol_fspai = "numeric",
definition="integer")
)
#'Constructor for inferenceDataObject class
#'
#'@param test A string defining the type of test to be performed. Multiple tests can be required. In this case the length of the list needs to be coherent with the ones of \code{type},
#' \code{component} and \code{interval}. The default is NULL, and can take values:
#'\itemize{
#'\item{'oat'}{: one-at-the-time tests, available only when \code{component} is 'parametric'.}
#'\item{'sim'}{: simultaneous tests.}
#'\item{'none'}{: no test required. \code{interval} must be set.}
#'}
#'@param interval A string defining the type of confidence intervals to be computed. Multiple intervals can be required. In this case the length of the list needs to be coherent with the ones of \code{type},
#'\code{component} and \code{test}. The default is NULL, and can take values:
#'\itemize{
#'\item{'oat'}{: one-at-the-time intervals.}
#'\item{'sim'}{: simultaneous intervals, available only when \code{component} is 'parametric' and no sign-flipping approaches are required.}
#'\item{'bonf'}{: Bonferroni intervals, available only when \code{component} is 'parametric'}
#'\item{'none'}{: no interval required. \code{test} must be set.}
#'}
#'@param type A list of strings defining the type of implementation for the inferential analysis. The possible values are:
#'\itemize{
#'\item{'w'}{: Wald parametric approach (default).}
#'\item{'s'}{: Speckman parametric approach.}
#'\item{'sf'}{: sign-flip nonparametric approach.}
#'\item{'esf'}{: eigen-sign-flip nonparametric approach.}
#'\item{'enh-esf'}{: enhanced-eigen-sign-flip nonparametric approach.}
#'}
#'@param component A list of strings defining on which model component inference has to be performed. It can take values 'parametric' (default), 'nonparametric' or 'both'.
#'@param dim Dimension of the problem, defaulted to NULL. It can take value 2 or 3 corresponding to 1.5D/2D or 2.5D/3D problems (Must be set by the user)
#'@param n_cov Number of the covariates, defaulted to NULL. (Must be set by the user)
#'@param locations A matrix of the locations of interest when testing the nonparametric component f, defaulted to NULL
#'@param locations_indices A vector of indices indicating the locations to be considered among the observed ones for nonparametric inference, defaulted to NULL.
#'If a vector of indices is provided, then the slot 'locations' is discarded.
#'@param locations_by_nodes A logical used to indicate wether the selected locations to perform inference on f are all coinciding with the nodes;
#'@param coeff A matrix, with \code{n_cov} number of columns, of numeric coefficients representing the linear combinations of the parametric components of the model.
#'The default is NULL, corresponding to an identity matrix. If at least one sing-flipping approach is required in \code{type}, needs to be an identity matrix.
#'@param beta0 Vector of real numbers (default NULL). It is used only if the \code{test} parameter is set, and \code{component} is not 'nonparametric'; its length is the number of rows of matrix \code{coeff} if provided.
#'If \code{test} is set and \code{beta0} is NULL, will be set to a vector of zeros.
#'@param f0 A function object representing the expression of the nonparametric component f under the null hypothesis. Taken into account if \code{test} is set and \code{component} is not parametric.
#'If NULL, the default is the null function, hence a test on the significance of the nonparametric component is carried out.
#'@param f_var A logical used to decide whether to estimate the local variance of the nonlinear part of the model.
#'The possible values are: FALSE (default) and TRUE.
#'@param level A vector containing the level of significance used to compute quantiles for confidence intervals, defaulted to 0.95. It is taken into account only if \code{interval} is set.
#'@param n_flip Number of flips performed in sign-flipping approaches, defaulted to 1000.
#'@return The output is a well defined \code{\link{inferenceDataObject}}, that can be used as input parameter in the \code{\link{smooth.FEM}} function.
#'@description A function that build an \code{\link{inferenceDataObject}}. In the process of construction many checks over the input parameters are carried out so that the output is a well defined object,
#'that can be used as parameter in \code{\link{smooth.FEM}} or \code{\link{smooth.FEM.time}} functions. Notice that this constructor ensures well-posedness of the object, but a further check on consistency with the smoothing functions parameters will be carried out.
#'@usage inferenceDataObjectBuilder(test = NULL,
#'interval = NULL,
#'type = 'w',
#'component = 'parametric',
#'dim = NULL,
#'n_cov = NULL,
#'locations = NULL,
#'locations_indices = NULL,
#'locations_by_nodes = FALSE,
#'coeff = NULL,
#'beta0 = NULL,
#'f0 = NULL,
#'f_var = FALSE,
#'level = 0.95,
#'n_flip = 1000)
#' @export
#' @examples
#' obj<-inferenceDataObjectBuilder(test = 'oat', dim = 2, beta0 = rep(1,4), n_cov = 4);
#' obj2<-inferenceDataObjectBuilder(test = 'sim', dim = 3, component = 'nonparametric', n_cov = 3);
inferenceDataObjectBuilder<-function(test = NULL,
interval = NULL,
type = "w",
component = "parametric",
dim = NULL,
n_cov = NULL,
locations = NULL,
locations_indices = NULL,
locations_by_nodes = FALSE,
coeff = NULL,
beta0 = NULL,
f0 = NULL,
f_var = FALSE,
level = 0.95,
n_flip = 1000)
{
# Preliminary check of parameters input types, translation into numeric representation of default occurrences.
if(!is.null(test)){
if(!is(test,"character"))
stop("'test'should be a character: choose one between 'oat', 'sim', 'none'")
if(length(test)==0)
stop("'test' is zero dimensional, should be a vector taking values among 'ont-at-the-time', 'sim', 'none'")
}
if(!is.null(interval)){
if(!is(interval,"character"))
stop("'interval'should be a character: choose one between 'oat' , 'sim', 'bonf', 'none'")
if(length(interval)==0)
stop("'interval' is zero dimensional, should be a vector taking values among 'oat', 'sim', 'bonf', 'none'")
}
if(length(type)==0)
stop("'type' is zero dimensional, should be a vector taking values among 'w', 's', 'sf' or 'esf'")
if(length(type) > 1 || type!="w"){
if(!is(type,"character"))
stop("'type' should be a vector of characters: choose among 'w', 's', 'sf' or 'esf'" )
}
if(length(component)==0)
stop("'component' is zero dimensional, should be a vector of characters taking values among 'parametric', 'nonparametric' or 'both'")
if(length(component) > 1 || component!="parametric"){
if(!is(component,"character"))
stop("'component' should be a vector of characters with values among 'parametric', 'nonparametric' or 'both'" )
}
if(!is.null(dim)){
if((!is(dim,"numeric")) & (!is(dim,"integer")))
stop("'dim' should be an integer or convertible to integer type")
else if(dim!=2 & dim!=3)
stop("'dim' should be either 2 or 3")
dim=as.integer(dim)
}
if(!is.null(n_cov)){
if((!is(n_cov,"numeric")) & (!is(n_cov,"integer")))
stop("'n_cov' should be an integer or convertible to integer type")
n_cov=as.integer(n_cov)
}
if(!is.null(locations)){
if(!is(locations,"matrix"))
stop("'locations' should be of class matrix")
}
if(!is.null(locations_indices)){
if((!is(locations_indices,"numeric")) & (!is(locations_indices,"integer")))
stop("'locations_indices' should be of class numeric or convertible to integer")
else{
for(i in 1:length(locations_indices)){
j <- as.integer(locations_indices[i])
if(locations_indices[i]!=j || locations_indices[i] <= 0)
stop("'locations_indices' should contain positive integers")
}
rm(list = c("i","j"))
}
}
if(locations_by_nodes != F){
if(!is(locations_by_nodes,"logical"))
stop("'locations_by_nodes' should be either TRUE or FALSE ")
if(length(locations_by_nodes)==0)
stop("'locations_by_nodes' is zero dimensional, should be either TRUE or FALSE")
if(locations_by_nodes!=T)
stop("'locations_by_nodes' should be either TRUE or FALSE")
locations_by_nodes_numeric=as.integer(1)
}else{
locations_by_nodes_numeric=as.integer(2)
}
if(!is.null(coeff)){
if(!is(coeff,"matrix"))
stop("'coeff' should be of class matrix")
}
if(!is.null(beta0)){
if(!is(beta0,"numeric"))
stop("'beta0' should be numeric")
if(length(beta0)==0)
stop("'beta0' is zerodimensional")
}
if(!is.null(f0)){
if(!is(f0,"function"))
stop("'f0' should be a function")
if(length(f0)==0)
stop("'f0' is zerodimensional")
}
if(f_var != F){
if(!is(f_var,"logical"))
stop("'f_var' should be either TRUE or FALSE ")
if(length(f_var)==0)
stop("'f_var' is zero dimensional, should be either TRUE or FALSE")
if(f_var!=T)
stop("'f_var' should be either TRUE or FALSE")
f_var_numeric=as.integer(1)
}else{
f_var_numeric=as.integer(2)
}
if(length(level)==0)
stop("'level' is zerodimensional, should be a vector of positive numbers between 0 and 1")
if(length(level) > 1 || level!=0.95){
if(!is(level,"numeric"))
stop("'level' should be numeric")
}
if(!is.null(n_flip)){
if((!is(n_flip,"numeric")) & (!is(n_flip,"integer")))
stop("'n_flip' should be an integer or convertible to integer type")
n_flip=as.integer(n_flip)
}
# Check of consistency of parameters. Translation into numeric representation. The checks are repeated for each element of the vectors test, interval and type
if(is.null(test) & is.null(interval)) # However, they can be both set
stop("at least one between 'test' and 'interval' should be not NULL")
if(sum(component == "parametric")!=length(component)){
if(is.null(dim))
stop("dimension of the problem is needed") # Otherwise the function won't be able to default f0
}
else{
if(is.null(dim))
dim = as.integer(0) # Inference on f is not required, we set dim to zero by default since it won't be required
}
if(sum(component == "nonparametric")!=length(component)){
if((is.null(n_cov) || n_cov <= 0)) # Otherwise the function won't be able to default the coeff neither object nor beta0
stop("number of covariates is needed")
else{
if(is.null(coeff)){ # If it is left as NULL, all the parameters are individually taken into account without any linear combination.
coeff = diag(1, nrow=n_cov, ncol=n_cov)
}
else{
if(dim(coeff)[2]!=n_cov)
stop("number of covariates and coefficients do not match")
for (i in 1:dim(coeff)[1]){
for(j in 1:dim(coeff)[2]){
if(!is(coeff[i,j],"numeric"))
stop("'coeff' should be composed by numeric values")
}
}
rm(list = ("i"))
rm(list = ("j"))
}
}
}else{
# inference on beta is not required, set n_cov and coeff to a 1x1 matrix
if(!is.null(n_cov) & n_cov >= 0){
n_cov = as.integer(n_cov)
}
else{
n_cov = as.integer(0)
}
coeff = matrix(data=0, nrow=1, ncol=1)
}
if(sum(component == "parametric")==length(component)){
if(!is.null(locations) || !is.null(locations_indices))
warning("locations are provided but not used, since inference is requested only on the paramteric component")
locations_indices = NULL
locations = matrix(data = 1, nrow = 1, ncol = 1)
}
else{
if(is.null(locations) & is.null(locations_indices)){
# default case: all observed locations are considered
locations <- matrix(data=1, nrow=1, ncol=1) # Just for convenience, the actual default will be set inside checkInferenceParameters
}
else if(is.null(locations_indices)){
if(sum(type == "esf")==1 || sum(type == "sf")==1)
stop("for the esf implementation a vector of location indices is required")
if(ncol(locations)!=dim)
stop("number of columns of 'locations' and 'dim' do not match")
}
else{
if(!is.null(locations))
warning("'locations' are provided but not used, since 'locations_indices' is not NULL")
# other dimensional checks on 'locations_indices' will be performed later, inside checkInferenceParameters
}
}
# Consistency check for number of implementations required (default values assigned in case of NULL)
n_of_implementations=max(length(type), length(test), length(interval), length(component));
# Pre-allocation of space
test_numeric=rep(0, n_of_implementations)
interval_numeric=rep(0, n_of_implementations)
component_numeric=rep(0, n_of_implementations)
type_numeric=rep(0, n_of_implementations)
quantile=rep(0, n_of_implementations)
if(!is.null(test)){
if(length(test) < n_of_implementations){
if(length(test) > 1)
stop("test vector length is not consistent with the number of implementations provided")
else
test = rep(test, n_of_implementations)
}
}else{test = rep("none", n_of_implementations)}
if(!is.null(interval)){
if(length(interval) < n_of_implementations){
if(length(interval) > 1)
stop("intervals vector length is not consistent with the number of implementations provided")
else
interval = rep(interval, n_of_implementations)
}
}else{interval = rep("none", n_of_implementations)}
if(!is.null(component)){
if(length(component) < n_of_implementations){
if(length(component) > 1)
stop("component vector length is not consistent with the number of implementations provided")
else
component = rep(component, n_of_implementations)
}
}
if(!is.null(type)){
if(length(type) < n_of_implementations){
if(length(type) > 1)
stop("type vector length is not consistent with the number of implementations provided")
else
type = rep(type, n_of_implementations)
}
}
# check level consistency
if(!is.null(level)){
n_of_intervals <- 0
for(i in 1:n_of_implementations){
if(interval[i]!="none")
n_of_intervals <- n_of_intervals + 1
}
if(length(level)==1){
level_vec <- rep(level, n_of_implementations)
alpha <- rep(1-level, n_of_implementations)
}
else if(length(level)==n_of_intervals){
level_vec <- numeric(n_of_implementations)
alpha <- numeric(n_of_implementations)
for(i in 1:n_of_implementations){
if(interval[i]!="none")
level_vec[i] <- level[i]
else
level_vec[i] <- 0
alpha[i] <- 1 - level_vec[i]
}
}
else
stop("level vector length is not consistent with the number of confidence interval implementations requested")
level = level_vec
rm(list = ("level_vec"))
rm(list = ("i"))
}
for (index in 1:n_of_implementations){
if(type[index]!="w" & type[index]!="s" & type[index]!="esf" & type[index]!="enh-esf" & type[index]!="sf"){
stop("type should be chosen between 'w', 's' 'sf', 'esf' or 'enh-esf'")}else{
if(type[index]=="w") type_numeric[index]=as.integer(1)
if(type[index]=="s") type_numeric[index]=as.integer(2)
if(type[index]=="esf") type_numeric[index]=as.integer(3)
if(type[index]=="enh-esf") type_numeric[index]=as.integer(4)
if(type[index]=="sf") type_numeric[index]=as.integer(5)
}
if(component[index]!="parametric" & component[index]!="nonparametric" & component[index]!="both"){
stop("component should be chosen between 'parametric', 'nonparametric' and 'both'")}else{
if(component[index]=="parametric") component_numeric[index]=as.integer(1)
if(component[index]=="nonparametric") component_numeric[index]=as.integer(2)
if(component[index]=="both") component_numeric[index]=as.integer(3)
}
if(type[index]=="sf" & component[index]!="nonparametric")
stop("sign-flip test is implemented only for the nonparametric component")
if(type[index]=="s" & component[index]!="parametric")
stop("speckman test and confidence intervals are implemented only for the parametric component")
if(test[index]=='none' & interval[index]=='none'){
stop("at least one between test and interval should be required for each implementation")
}
if(test[index]!="oat" & test[index]!="sim" & test[index]!="none"){
stop("test should be either 'oat', 'sim' or 'none'")}else{
if(test[index]=="none") test_numeric[index]=as.integer(0)
if(test[index]=="oat") test_numeric[index]=as.integer(1)
if(test[index]=="sim") test_numeric[index]=as.integer(2)
}
if(interval[index]!="none"){
if(interval[index]!="oat" & interval[index]!="sim" & interval[index]!="bonf" & interval[index]!="none"){
stop("interval should be either 'oat' 'sim', 'bonf' or 'none'")}else{
if(interval[index]=="none") interval_numeric[index]=as.integer(0)
if(interval[index]=="oat") interval_numeric[index]=as.integer(1)
if(interval[index]=="sim") interval_numeric[index]=as.integer(2)
if(interval[index]=="bonf") interval_numeric[index]=as.integer(3)
}
if(level[index] <= 0 || level[index] > 1)
stop("level should be a positive value smaller or equal to 1")
}
if(interval[index]=="sim" & (type[index]=="esf" || type[index] == "enh-esf")){
stop("simultaneous confidence intervals are not implemented with eigen-sign-flip implementation")
}
if((interval[index]=="sim" || interval[index]=="bonf") & component[index]!="parametric")
stop("only one-at-the-time confidence intervals are allowed for the nonparametric component")
if((type[index] == "esf" || type[index] == "enh-esf") & component[index]!="nonparametric"){
for(i in 1:dim(coeff)[1]){
count=0
for(j in 1:dim(coeff)[2]){
if(coeff[i,j]!= 0 & coeff[i,j]!=1)
stop("linear combinations are not allowed in the eigen-sign-flip cases")
count = count + coeff[i,j]
}
if(count != 1)
stop("linear combinations are not allowed in the eigen-sign-flip cases")
}
}
# Well posedness check for coeff in simultaneous case;
if(test[index]=="sim" & component[index]!="nonparametric"){
if(det(coeff %*% t(coeff)) < 0.001){
stop("coeff is not full rank, variance-covariance matrix of the linear combination not invertible")
}
}
if(interval[index]=="sim" & component[index]!="nonparametric"){
if(det(coeff %*% t(coeff)) < 0.001){
stop("coeff is not full rank, variance-covariance matrix of the linear combination not invertible")
}
}
if((test[index]=="oat") & component[index] != "parametric"){
warning("only simultaneous tests are available for the nonparametric component, proceeding with simultaneous inference")
test_numeric[index] = as.integer(2)
}
# Build the quantile for confidence intervals if needed
if(interval[index]=="none" || component[index]=="nonparametric"){
quantile[index]=0
}else{
if(level[index] > 0){
if(interval[index] == "sim"){ # Simultaneous CI -> Chi-Squared (q) law for statistic
quantile[index]=sqrt(qchisq(level[index], dim(coeff)[1]))
}else{
if(interval[index]=="oat"){ # One at the time CI (each interval has confidence alpha) -> Gaussian law for statistic
quantile[index]=qnorm((1-(1-level[index])/2),0,1)
}else{
if(interval[index]== "bonf"){# One at the time CI (overall confidence alpha) -> Gaussian law for statistic
quantile[index]=qnorm((1-(1-level[index])/(2*dim(coeff)[1])),0,1)
}
}
}
}
}
} # End of for cycle over the different implementation
if(sum(component == "nonparametric")!=length(component)){
if(is.null(beta0)) # If left to NULL, beta0 is set to a vector of zeros.
beta0<-rep(0, dim(coeff)[1])
else{
if(length(beta0)!=dim(coeff)[1])
stop("dimension of 'coeff' and 'beta0' are not consistent")
}
}else{
# inference on beta is not required, just setting it to zero
beta0<-0
}
if(sum(component == "parametric")!=length(component)){
if(is.null(f0)){ # If left to NULL, f0 is set to a null function.
f0_3D <- function (x,y,z){return(0)}
}
else{
args<-formals(f0)
non_defaulted_args = 0
for(i in 1:length(args)){
if(is.name(args[[i]]))
non_defaulted_args <- non_defaulted_args + 1
}
if(non_defaulted_args != dim)
stop("number of f0 coordinate arguments is not consistent with the problem dimension 'dim'")
rm(list = c("non_defaulted_args", "i"))
if(dim==2){ # We need to upgrade the function to 3D definition for documentation
f0_3D <- function(x,y,z){return(f0(x,y))}
}else{ # We are already in 3D case
f0_3D <- f0
}
}
}
else{
# inference on the nonparametric component is not required, just setting f0 to a zero function
f0_3D <- function(x,y,z){return(0)}
}
if(!is.null(n_flip)){
if(n_flip <= 0)
stop("number of sign-flips must be a positive value")
}else {
n_flip <- as.integer(1000)
}
definition=as.integer(1)
# Building the output object, returning it
if(!is.null(locations_indices))
result<-new("inferenceDataObject", test = as.integer(test_numeric), interval = as.integer(interval_numeric), type = as.integer(type_numeric), component = as.integer(component_numeric), exact = as.integer(1), dim = dim, n_cov = n_cov,
locations_indices = as.integer(locations_indices), locations_are_nodes = locations_by_nodes_numeric, coeff = coeff, beta0 = beta0, f0 = f0_3D, f_var = f_var_numeric, quantile = quantile, alpha = alpha, n_flip = n_flip, tol_fspai = 0.05, definition=definition)
else
result<-new("inferenceDataObject", test = as.integer(test_numeric), interval = as.integer(interval_numeric), type = as.integer(type_numeric), component = as.integer(component_numeric), exact = as.integer(1), dim = dim, n_cov = n_cov,
locations = locations, locations_are_nodes =locations_by_nodes_numeric, coeff = coeff, beta0 = beta0, f0 = f0_3D, f_var = f_var_numeric, quantile = quantile, alpha = alpha, n_flip = n_flip, tol_fspai = 0.05, definition=definition)
return(result)
}
Try the fdaPDE package in your browser
Any scripts or data that you put into this service are public.
fdaPDE documentation built on March 7, 2023, 5:28 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Defines functions inferenceDataObjectBuilder
Documented in inferenceDataObjectBuilder
#' @useDynLib fdaPDE
#' @importFrom methods new
#' @importFrom stats pchisq pnorm qchisq qnorm
NULL
#' Class for inference data
#'
#'@slot test A vector of integers taking value 0, 1 or 2; if 0 no test is performed, if 1 one-at-the-time tests are performed, if 2 a simultaneous test is performed.
#'@slot interval A vector of integers taking value 0, 1, 2 or 3; if 0 no confidence interval is computed, if 1 one-at-the-time confidence intervals are computed,
#'if 2 simultaneous confidence intervals are computed, if 3 Bonferroni confidence intervals are computed.
#'@slot type A vector of integers taking value 1, 2, 3, 4 or 5 corresponding to Wald, Speckman, Eigen-Sign-Flip, Enhanced-Eigen-Sign-Flip or Sign-Flip inferential approach.
#'@slot component A vector of integers taking value 1, 2 or 3, indicating whether the inferential analysis should be carried out respectively for the parametric, nonparametric or both the components.
#'@slot exact An integer taking value 1 or 2. If 1 an exact computation of the test statistics will be performed,
#'whereas if 2 an approximated computation will be carried out (not implemented in this version).
#'@slot dim Dimension of the problem, it is equal to 2 in the 1.5D and 2D cases and equal to 3 in the 2.5D and 3D cases.
#'@slot n_cov Number of covariates taken into account in the linear part of the regression problem.
#'@slot locations A matrix of numeric coefficients with columns of dimension \code{dim}. When nonparametric inference is requested it represents the set of spatial locations for which the inferential analysis should be performed.
#' The default values is a one-dimensional matrix of value 1 indicating that all the observed location points should be considered.
#' In the sign-flip and eigen-sign-flip implementations only observed points are allowed.
#'@slot locations_indices A vector of indices indicating which spatial points have to be considered among the observed ones for nonparametric inference. If a vector of location indices is provided
#'then the slot 'location' is discarded.
#'@slot locations_are_nodes An integer taking value 1 or 2; in the first case it indicates that the selected locations to perform inference on f are all coinciding with the nodes; otherwise it takes value 2;
#'@slot coeff A matrix of numeric coefficients with columns of dimension \code{n_cov} and each row represents a linear combination of the linear parameters to be tested and/or to be
#' estimated via confidence interval.
#'@slot beta0 Vector of null hypothesis values for the linear parameters of the model. Used only if \code{test} is not 0 and \code{component} is not 2.
#'@slot f0 Function representing the expression of the nonparametric component f under the null hypothesis. Used only if \code{component} is not 1.
#'@slot f0_eval Vector of f0 evaluations at the choosen test locations. It will be eventually set later in checkInferenceParameters, if nonparametric inference is required.
#'@slot f_var An integer taking value 1 or 2. If 1 local variance estimates for the nonlinear part of the model will be computed,
#'whereas if 2 they will not.
#'@slot quantile Vector of quantiles needed for confidence intervals, used only if interval is not 0.
#'@slot alpha 1 minus confidence level vector of sign-flipping approaches confidence intervals. Used only if interval is not 0.
#'@slot n_flip An integer representing the number of sign-flips in the case of sign-flipping approaches.
#'@slot tol_fspai A real number greater than 0 specifying the tolerance for FSPAI algorithm, in case of non-exact inference (not implemented in this version).
#'@slot definition An integer taking value 0 or 1. If set to 1, the class will be considered as created by the function \code{\link{inferenceDataObjectBuilder}},
#'leading to avoid some of the checks that are performed on inference data within smoothing functions.
#'
#'@description
#' A class that contains all possible information for inference over linear parameters and/or nonparametric field in spatial regression with
#' differential regularization problem. This object can be used as parameter in smoothing function of the fdaPDE library \code{\link{smooth.FEM}}.
#'
#'@details
#'At least one between test and interval must be nonzero. \code{n_cov}, \code{coeff} and \code{beta0}, if provided, need to be coherent.
#'\code{dim} and \code{locations}, if provided, need to be coherent.
#'The usage of \code{\link{inferenceDataObjectBuilder}} is recommended for the construction of an object of this class.
inferenceDataObject<-setClass("inferenceDataObject", slots = list(test = "integer",
interval = "integer",
type = "integer",
component = "integer",
exact = "integer",
dim = "integer",
n_cov = "integer",
locations = "matrix",
locations_indices = "integer",
locations_are_nodes = "integer",
coeff = "matrix",
beta0 = "numeric",
f0 = "function",
f0_eval = "numeric",
f_var = "integer",
quantile = "numeric",
alpha = "numeric",
n_flip = "integer",
tol_fspai = "numeric",
definition="integer")
)
#'Constructor for inferenceDataObject class
#'
#'@param test A string defining the type of test to be performed. Multiple tests can be required. In this case the length of the list needs to be coherent with the ones of \code{type},
#' \code{component} and \code{interval}. The default is NULL, and can take values:
#'\itemize{
#'\item{'oat'}{: one-at-the-time tests, available only when \code{component} is 'parametric'.}
#'\item{'sim'}{: simultaneous tests.}
#'\item{'none'}{: no test required. \code{interval} must be set.}
#'}
#'@param interval A string defining the type of confidence intervals to be computed. Multiple intervals can be required. In this case the length of the list needs to be coherent with the ones of \code{type},
#'\code{component} and \code{test}. The default is NULL, and can take values:
#'\itemize{
#'\item{'oat'}{: one-at-the-time intervals.}
#'\item{'sim'}{: simultaneous intervals, available only when \code{component} is 'parametric' and no sign-flipping approaches are required.}
#'\item{'bonf'}{: Bonferroni intervals, available only when \code{component} is 'parametric'}
#'\item{'none'}{: no interval required. \code{test} must be set.}
#'}
#'@param type A list of strings defining the type of implementation for the inferential analysis. The possible values are:
#'\itemize{
#'\item{'w'}{: Wald parametric approach (default).}
#'\item{'s'}{: Speckman parametric approach.}
#'\item{'sf'}{: sign-flip nonparametric approach.}
#'\item{'esf'}{: eigen-sign-flip nonparametric approach.}
#'\item{'enh-esf'}{: enhanced-eigen-sign-flip nonparametric approach.}
#'}
#'@param component A list of strings defining on which model component inference has to be performed. It can take values 'parametric' (default), 'nonparametric' or 'both'.
#'@param dim Dimension of the problem, defaulted to NULL. It can take value 2 or 3 corresponding to 1.5D/2D or 2.5D/3D problems (Must be set by the user)
#'@param n_cov Number of the covariates, defaulted to NULL. (Must be set by the user)
#'@param locations A matrix of the locations of interest when testing the nonparametric component f, defaulted to NULL
#'@param locations_indices A vector of indices indicating the locations to be considered among the observed ones for nonparametric inference, defaulted to NULL.
#'If a vector of indices is provided, then the slot 'locations' is discarded.
#'@param locations_by_nodes A logical used to indicate wether the selected locations to perform inference on f are all coinciding with the nodes;
#'@param coeff A matrix, with \code{n_cov} number of columns, of numeric coefficients representing the linear combinations of the parametric components of the model.
#'The default is NULL, corresponding to an identity matrix. If at least one sing-flipping approach is required in \code{type}, needs to be an identity matrix.
#'@param beta0 Vector of real numbers (default NULL). It is used only if the \code{test} parameter is set, and \code{component} is not 'nonparametric'; its length is the number of rows of matrix \code{coeff} if provided.
#'If \code{test} is set and \code{beta0} is NULL, will be set to a vector of zeros.
#'@param f0 A function object representing the expression of the nonparametric component f under the null hypothesis. Taken into account if \code{test} is set and \code{component} is not parametric.
#'If NULL, the default is the null function, hence a test on the significance of the nonparametric component is carried out.
#'@param f_var A logical used to decide whether to estimate the local variance of the nonlinear part of the model.
#'The possible values are: FALSE (default) and TRUE.
#'@param level A vector containing the level of significance used to compute quantiles for confidence intervals, defaulted to 0.95. It is taken into account only if \code{interval} is set.
#'@param n_flip Number of flips performed in sign-flipping approaches, defaulted to 1000.
#'@return The output is a well defined \code{\link{inferenceDataObject}}, that can be used as input parameter in the \code{\link{smooth.FEM}} function.
#'@description A function that build an \code{\link{inferenceDataObject}}. In the process of construction many checks over the input parameters are carried out so that the output is a well defined object,
#'that can be used as parameter in \code{\link{smooth.FEM}} or \code{\link{smooth.FEM.time}} functions. Notice that this constructor ensures well-posedness of the object, but a further check on consistency with the smoothing functions parameters will be carried out.
#'@usage inferenceDataObjectBuilder(test = NULL,
#'interval = NULL,
#'type = 'w',
#'component = 'parametric',
#'dim = NULL,
#'n_cov = NULL,
#'locations = NULL,
#'locations_indices = NULL,
#'locations_by_nodes = FALSE,
#'coeff = NULL,
#'beta0 = NULL,
#'f0 = NULL,
#'f_var = FALSE,
#'level = 0.95,
#'n_flip = 1000)
#' @export
#' @examples
#' obj<-inferenceDataObjectBuilder(test = 'oat', dim = 2, beta0 = rep(1,4), n_cov = 4);
#' obj2<-inferenceDataObjectBuilder(test = 'sim', dim = 3, component = 'nonparametric', n_cov = 3);
inferenceDataObjectBuilder<-function(test = NULL,
interval = NULL,
type = "w",
component = "parametric",
dim = NULL,
n_cov = NULL,
locations = NULL,
locations_indices = NULL,
locations_by_nodes = FALSE,
coeff = NULL,
beta0 = NULL,
f0 = NULL,
f_var = FALSE,
level = 0.95,
n_flip = 1000)
{
# Preliminary check of parameters input types, translation into numeric representation of default occurrences.
if(!is.null(test)){
if(!is(test,"character"))
stop("'test'should be a character: choose one between 'oat', 'sim', 'none'")
if(length(test)==0)
stop("'test' is zero dimensional, should be a vector taking values among 'ont-at-the-time', 'sim', 'none'")
}
if(!is.null(interval)){
if(!is(interval,"character"))
stop("'interval'should be a character: choose one between 'oat' , 'sim', 'bonf', 'none'")
if(length(interval)==0)
stop("'interval' is zero dimensional, should be a vector taking values among 'oat', 'sim', 'bonf', 'none'")
}
if(length(type)==0)
stop("'type' is zero dimensional, should be a vector taking values among 'w', 's', 'sf' or 'esf'")
if(length(type) > 1 || type!="w"){
if(!is(type,"character"))
stop("'type' should be a vector of characters: choose among 'w', 's', 'sf' or 'esf'" )
}
if(length(component)==0)
stop("'component' is zero dimensional, should be a vector of characters taking values among 'parametric', 'nonparametric' or 'both'")
if(length(component) > 1 || component!="parametric"){
if(!is(component,"character"))
stop("'component' should be a vector of characters with values among 'parametric', 'nonparametric' or 'both'" )
}
if(!is.null(dim)){
if((!is(dim,"numeric")) & (!is(dim,"integer")))
stop("'dim' should be an integer or convertible to integer type")
else if(dim!=2 & dim!=3)
stop("'dim' should be either 2 or 3")
dim=as.integer(dim)
}
if(!is.null(n_cov)){
if((!is(n_cov,"numeric")) & (!is(n_cov,"integer")))
stop("'n_cov' should be an integer or convertible to integer type")
n_cov=as.integer(n_cov)
}
if(!is.null(locations)){
if(!is(locations,"matrix"))
stop("'locations' should be of class matrix")
}
if(!is.null(locations_indices)){
if((!is(locations_indices,"numeric")) & (!is(locations_indices,"integer")))
stop("'locations_indices' should be of class numeric or convertible to integer")
else{
for(i in 1:length(locations_indices)){
j <- as.integer(locations_indices[i])
if(locations_indices[i]!=j || locations_indices[i] <= 0)
stop("'locations_indices' should contain positive integers")
}
rm(list = c("i","j"))
}
}
if(locations_by_nodes != F){
if(!is(locations_by_nodes,"logical"))
stop("'locations_by_nodes' should be either TRUE or FALSE ")
if(length(locations_by_nodes)==0)
stop("'locations_by_nodes' is zero dimensional, should be either TRUE or FALSE")
if(locations_by_nodes!=T)
stop("'locations_by_nodes' should be either TRUE or FALSE")
locations_by_nodes_numeric=as.integer(1)
}else{
locations_by_nodes_numeric=as.integer(2)
}
if(!is.null(coeff)){
if(!is(coeff,"matrix"))
stop("'coeff' should be of class matrix")
}
if(!is.null(beta0)){
if(!is(beta0,"numeric"))
stop("'beta0' should be numeric")
if(length(beta0)==0)
stop("'beta0' is zerodimensional")
}
if(!is.null(f0)){
if(!is(f0,"function"))
stop("'f0' should be a function")
if(length(f0)==0)
stop("'f0' is zerodimensional")
}
if(f_var != F){
if(!is(f_var,"logical"))
stop("'f_var' should be either TRUE or FALSE ")
if(length(f_var)==0)
stop("'f_var' is zero dimensional, should be either TRUE or FALSE")
if(f_var!=T)
stop("'f_var' should be either TRUE or FALSE")
f_var_numeric=as.integer(1)
}else{
f_var_numeric=as.integer(2)
}
if(length(level)==0)
stop("'level' is zerodimensional, should be a vector of positive numbers between 0 and 1")
if(length(level) > 1 || level!=0.95){
if(!is(level,"numeric"))
stop("'level' should be numeric")
}
if(!is.null(n_flip)){
if((!is(n_flip,"numeric")) & (!is(n_flip,"integer")))
stop("'n_flip' should be an integer or convertible to integer type")
n_flip=as.integer(n_flip)
}
# Check of consistency of parameters. Translation into numeric representation. The checks are repeated for each element of the vectors test, interval and type
if(is.null(test) & is.null(interval)) # However, they can be both set
stop("at least one between 'test' and 'interval' should be not NULL")
if(sum(component == "parametric")!=length(component)){
if(is.null(dim))
stop("dimension of the problem is needed") # Otherwise the function won't be able to default f0
}
else{
if(is.null(dim))
dim = as.integer(0) # Inference on f is not required, we set dim to zero by default since it won't be required
}
if(sum(component == "nonparametric")!=length(component)){
if((is.null(n_cov) || n_cov <= 0)) # Otherwise the function won't be able to default the coeff neither object nor beta0
stop("number of covariates is needed")
else{
if(is.null(coeff)){ # If it is left as NULL, all the parameters are individually taken into account without any linear combination.
coeff = diag(1, nrow=n_cov, ncol=n_cov)
}
else{
if(dim(coeff)[2]!=n_cov)
stop("number of covariates and coefficients do not match")
for (i in 1:dim(coeff)[1]){
for(j in 1:dim(coeff)[2]){
if(!is(coeff[i,j],"numeric"))
stop("'coeff' should be composed by numeric values")
}
}
rm(list = ("i"))
rm(list = ("j"))
}
}
}else{
# inference on beta is not required, set n_cov and coeff to a 1x1 matrix
if(!is.null(n_cov) & n_cov >= 0){
n_cov = as.integer(n_cov)
}
else{
n_cov = as.integer(0)
}
coeff = matrix(data=0, nrow=1, ncol=1)
}
if(sum(component == "parametric")==length(component)){
if(!is.null(locations) || !is.null(locations_indices))
warning("locations are provided but not used, since inference is requested only on the paramteric component")
locations_indices = NULL
locations = matrix(data = 1, nrow = 1, ncol = 1)
}
else{
if(is.null(locations) & is.null(locations_indices)){
# default case: all observed locations are considered
locations <- matrix(data=1, nrow=1, ncol=1) # Just for convenience, the actual default will be set inside checkInferenceParameters
}
else if(is.null(locations_indices)){
if(sum(type == "esf")==1 || sum(type == "sf")==1)
stop("for the esf implementation a vector of location indices is required")
if(ncol(locations)!=dim)
stop("number of columns of 'locations' and 'dim' do not match")
}
else{
if(!is.null(locations))
warning("'locations' are provided but not used, since 'locations_indices' is not NULL")
# other dimensional checks on 'locations_indices' will be performed later, inside checkInferenceParameters
}
}
# Consistency check for number of implementations required (default values assigned in case of NULL)
n_of_implementations=max(length(type), length(test), length(interval), length(component));
# Pre-allocation of space
test_numeric=rep(0, n_of_implementations)
interval_numeric=rep(0, n_of_implementations)
component_numeric=rep(0, n_of_implementations)
type_numeric=rep(0, n_of_implementations)
quantile=rep(0, n_of_implementations)
if(!is.null(test)){
if(length(test) < n_of_implementations){
if(length(test) > 1)
stop("test vector length is not consistent with the number of implementations provided")
else
test = rep(test, n_of_implementations)
}
}else{test = rep("none", n_of_implementations)}
if(!is.null(interval)){
if(length(interval) < n_of_implementations){
if(length(interval) > 1)
stop("intervals vector length is not consistent with the number of implementations provided")
else
interval = rep(interval, n_of_implementations)
}
}else{interval = rep("none", n_of_implementations)}
if(!is.null(component)){
if(length(component) < n_of_implementations){
if(length(component) > 1)
stop("component vector length is not consistent with the number of implementations provided")
else
component = rep(component, n_of_implementations)
}
}
if(!is.null(type)){
if(length(type) < n_of_implementations){
if(length(type) > 1)
stop("type vector length is not consistent with the number of implementations provided")
else
type = rep(type, n_of_implementations)
}
}
# check level consistency
if(!is.null(level)){
n_of_intervals <- 0
for(i in 1:n_of_implementations){
if(interval[i]!="none")
n_of_intervals <- n_of_intervals + 1
}
if(length(level)==1){
level_vec <- rep(level, n_of_implementations)
alpha <- rep(1-level, n_of_implementations)
}
else if(length(level)==n_of_intervals){
level_vec <- numeric(n_of_implementations)
alpha <- numeric(n_of_implementations)
for(i in 1:n_of_implementations){
if(interval[i]!="none")
level_vec[i] <- level[i]
else
level_vec[i] <- 0
alpha[i] <- 1 - level_vec[i]
}
}
else
stop("level vector length is not consistent with the number of confidence interval implementations requested")
level = level_vec
rm(list = ("level_vec"))
rm(list = ("i"))
}
for (index in 1:n_of_implementations){
if(type[index]!="w" & type[index]!="s" & type[index]!="esf" & type[index]!="enh-esf" & type[index]!="sf"){
stop("type should be chosen between 'w', 's' 'sf', 'esf' or 'enh-esf'")}else{
if(type[index]=="w") type_numeric[index]=as.integer(1)
if(type[index]=="s") type_numeric[index]=as.integer(2)
if(type[index]=="esf") type_numeric[index]=as.integer(3)
if(type[index]=="enh-esf") type_numeric[index]=as.integer(4)
if(type[index]=="sf") type_numeric[index]=as.integer(5)
}
if(component[index]!="parametric" & component[index]!="nonparametric" & component[index]!="both"){
stop("component should be chosen between 'parametric', 'nonparametric' and 'both'")}else{
if(component[index]=="parametric") component_numeric[index]=as.integer(1)
if(component[index]=="nonparametric") component_numeric[index]=as.integer(2)
if(component[index]=="both") component_numeric[index]=as.integer(3)
}
if(type[index]=="sf" & component[index]!="nonparametric")
stop("sign-flip test is implemented only for the nonparametric component")
if(type[index]=="s" & component[index]!="parametric")
stop("speckman test and confidence intervals are implemented only for the parametric component")
if(test[index]=='none' & interval[index]=='none'){
stop("at least one between test and interval should be required for each implementation")
}
if(test[index]!="oat" & test[index]!="sim" & test[index]!="none"){
stop("test should be either 'oat', 'sim' or 'none'")}else{
if(test[index]=="none") test_numeric[index]=as.integer(0)
if(test[index]=="oat") test_numeric[index]=as.integer(1)
if(test[index]=="sim") test_numeric[index]=as.integer(2)
}
if(interval[index]!="none"){
if(interval[index]!="oat" & interval[index]!="sim" & interval[index]!="bonf" & interval[index]!="none"){
stop("interval should be either 'oat' 'sim', 'bonf' or 'none'")}else{
if(interval[index]=="none") interval_numeric[index]=as.integer(0)
if(interval[index]=="oat") interval_numeric[index]=as.integer(1)
if(interval[index]=="sim") interval_numeric[index]=as.integer(2)
if(interval[index]=="bonf") interval_numeric[index]=as.integer(3)
}
if(level[index] <= 0 || level[index] > 1)
stop("level should be a positive value smaller or equal to 1")
}
if(interval[index]=="sim" & (type[index]=="esf" || type[index] == "enh-esf")){
stop("simultaneous confidence intervals are not implemented with eigen-sign-flip implementation")
}
if((interval[index]=="sim" || interval[index]=="bonf") & component[index]!="parametric")
stop("only one-at-the-time confidence intervals are allowed for the nonparametric component")
if((type[index] == "esf" || type[index] == "enh-esf") & component[index]!="nonparametric"){
for(i in 1:dim(coeff)[1]){
count=0
for(j in 1:dim(coeff)[2]){
if(coeff[i,j]!= 0 & coeff[i,j]!=1)
stop("linear combinations are not allowed in the eigen-sign-flip cases")
count = count + coeff[i,j]
}
if(count != 1)
stop("linear combinations are not allowed in the eigen-sign-flip cases")
}
}
# Well posedness check for coeff in simultaneous case;
if(test[index]=="sim" & component[index]!="nonparametric"){
if(det(coeff %*% t(coeff)) < 0.001){
stop("coeff is not full rank, variance-covariance matrix of the linear combination not invertible")
}
}
if(interval[index]=="sim" & component[index]!="nonparametric"){
if(det(coeff %*% t(coeff)) < 0.001){
stop("coeff is not full rank, variance-covariance matrix of the linear combination not invertible")
}
}
if((test[index]=="oat") & component[index] != "parametric"){
warning("only simultaneous tests are available for the nonparametric component, proceeding with simultaneous inference")
test_numeric[index] = as.integer(2)
}
# Build the quantile for confidence intervals if needed
if(interval[index]=="none" || component[index]=="nonparametric"){
quantile[index]=0
}else{
if(level[index] > 0){
if(interval[index] == "sim"){ # Simultaneous CI -> Chi-Squared (q) law for statistic
quantile[index]=sqrt(qchisq(level[index], dim(coeff)[1]))
}else{
if(interval[index]=="oat"){ # One at the time CI (each interval has confidence alpha) -> Gaussian law for statistic
quantile[index]=qnorm((1-(1-level[index])/2),0,1)
}else{
if(interval[index]== "bonf"){# One at the time CI (overall confidence alpha) -> Gaussian law for statistic
quantile[index]=qnorm((1-(1-level[index])/(2*dim(coeff)[1])),0,1)
}
}
}
}
}
} # End of for cycle over the different implementation
if(sum(component == "nonparametric")!=length(component)){
if(is.null(beta0)) # If left to NULL, beta0 is set to a vector of zeros.
beta0<-rep(0, dim(coeff)[1])
else{
if(length(beta0)!=dim(coeff)[1])
stop("dimension of 'coeff' and 'beta0' are not consistent")
}
}else{
# inference on beta is not required, just setting it to zero
beta0<-0
}
if(sum(component == "parametric")!=length(component)){
if(is.null(f0)){ # If left to NULL, f0 is set to a null function.
f0_3D <- function (x,y,z){return(0)}
}
else{
args<-formals(f0)
non_defaulted_args = 0
for(i in 1:length(args)){
if(is.name(args[[i]]))
non_defaulted_args <- non_defaulted_args + 1
}
if(non_defaulted_args != dim)
stop("number of f0 coordinate arguments is not consistent with the problem dimension 'dim'")
rm(list = c("non_defaulted_args", "i"))
if(dim==2){ # We need to upgrade the function to 3D definition for documentation
f0_3D <- function(x,y,z){return(f0(x,y))}
}else{ # We are already in 3D case
f0_3D <- f0
}
}
}
else{
# inference on the nonparametric component is not required, just setting f0 to a zero function
f0_3D <- function(x,y,z){return(0)}
}
if(!is.null(n_flip)){
if(n_flip <= 0)
stop("number of sign-flips must be a positive value")
}else {
n_flip <- as.integer(1000)
}
definition=as.integer(1)
# Building the output object, returning it
if(!is.null(locations_indices))
result<-new("inferenceDataObject", test = as.integer(test_numeric), interval = as.integer(interval_numeric), type = as.integer(type_numeric), component = as.integer(component_numeric), exact = as.integer(1), dim = dim, n_cov = n_cov,
locations_indices = as.integer(locations_indices), locations_are_nodes = locations_by_nodes_numeric, coeff = coeff, beta0 = beta0, f0 = f0_3D, f_var = f_var_numeric, quantile = quantile, alpha = alpha, n_flip = n_flip, tol_fspai = 0.05, definition=definition)
else
result<-new("inferenceDataObject", test = as.integer(test_numeric), interval = as.integer(interval_numeric), type = as.integer(type_numeric), component = as.integer(component_numeric), exact = as.integer(1), dim = dim, n_cov = n_cov,
locations = locations, locations_are_nodes =locations_by_nodes_numeric, coeff = coeff, beta0 = beta0, f0 = f0_3D, f_var = f_var_numeric, quantile = quantile, alpha = alpha, n_flip = n_flip, tol_fspai = 0.05, definition=definition)
return(result)
}
Try the fdaPDE package in your browser
Any scripts or data that you put into this service are public.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.