updated package function for model robust data generation and model m…

…isspecified subsampling
Amalan-ConStat · Oct 3, 2024 · d68d72d · d68d72d
1 parent 76720a5
commit d68d72d
Show file tree

Hide file tree

Showing 65 changed files with 1,139 additions and 1,627 deletions.
diff --git a/NAMESPACE b/NAMESPACE
@@ -13,7 +13,6 @@ export(AoptimalGauLMSub)
 export(AoptimalMCGLMSub)
 export(GenGLMdata)
 export(GenModelMissGLMdata)
-export(GenModelRobustGLMdata)
 export(LCCsampling)
 export(LeverageSampling)
 export(modelMissLinSub)

diff --git a/R/ALoptimalGLMSub.R b/R/ALoptimalGLMSub.R
@@ -465,182 +465,3 @@ ALoptimalGLMSub <- function(r1,r2,Y,X,N,family){
     return(ans)
   }
 }
-
-#' Generate data for Generalised Linear Models
-#'
-#' Function to simulate big data under linear, logistic and Poisson regression for sampling.
-#' Covariate data X is through Normal or Uniform distribution for linear regression.
-#' Covariate data X is through Exponential or Normal or Uniform distribution for logistic regression.
-#' Covariate data X is through Normal or Uniform distribution for Poisson regression.
-#'
-#' @usage
-#' GenGLMdata(Dist,Dist_Par,No_Of_Var,Beta,N,family)
-#'
-#' @param Dist        a character value for the distribution "Normal" or "Uniform or "Exponential"
-#' @param Dist_Par    a list of parameters for the distribution that would generate data for covariate X
-#' @param No_Of_Var   number of variables
-#' @param Beta        a vector for the model parameters, including the intercept
-#' @param N           the big data size
-#' @param family      a character vector for "linear", "logistic" and "poisson" regression from Generalised Linear Models
-#'
-#' @details
-#' Big data for the Generalised Linear Models are generated by the "linear", "logistic" and "poisson"
-#' regression types.
-#'
-#' We have limited the covariate data generation for
-#' linear regression through normal and uniform distribution,
-#' logistic regression through exponential, normal and uniform and
-#' Poisson regression through normal and uniform distribution.
-#'
-#' @return
-#' The output of \code{GenGLMData} gives a list of
-#'
-#' \code{Basic} a list of outputs based on the inputs and Beta Estimates for all models
-#'
-#' \code{Complete_Data} a matrix for Y and X
-#'
-#' @references
-#' \insertRef{lee1996hierarchical}{NeEDS4BigData}
-#'
-#' @examples
-#' Dist<-"Normal"; Dist_Par<-list(Mean=0,Variance=1,Error_Variance=0.5)
-#' No_Of_Var<-2; Beta<-c(-1,2,1); N<-5000; Family<-"linear"
-#' Results<-GenGLMdata(Dist,Dist_Par,No_Of_Var,Beta,N,Family)
-#'
-#' Dist<-"Normal"; Dist_Par<-list(Mean=0,Variance=1)
-#' No_Of_Var<-2; Beta<-c(-1,2,1); N<-5000; Family<-"logistic"
-#' Results<-GenGLMdata(Dist,Dist_Par,No_Of_Var,Beta,N,Family)
-#'
-#' Dist<-"Normal";
-#' No_Of_Var<-2; Beta<-c(-1,2,1); N<-5000; Family<-"poisson"
-#' Results<-GenGLMdata(Dist,NULL,No_Of_Var,Beta,N,Family)
-#'
-#' @import stats
-#' @export
-GenGLMdata<-function(Dist,Dist_Par,No_Of_Var,Beta,N,family){
-  if(any(is.na(c(Dist,Beta,No_Of_Var,N,family))) | any(is.nan(c(Dist,No_Of_Var,Beta,N,family)))){
-    stop("NA or Infinite or NAN values in the Dist,Beta,No_Of_Var,N or family")
-  }
-
-  if(any(is.na(unlist(Dist_Par))) | any(is.nan(unlist(Dist_Par)))){
-    stop("NA or Infinite or NAN values in the Dist_Par")
-  }
-
-  if(!any(family == c("linear","logistic","poisson"))){
-    stop("Only the regression types 'linear','logistic' or 'poisson' are allowed")
-  }
-
-  if(family == "linear"){
-    if(!(Dist == "Normal" | Dist == "Uniform")){
-      stop("For linear regression select the distribution 'Normal' \n or 'Uniform' to generate the covarate data")
-    }
-  }
-
-  if(family == "logistic"){
-    if(!(Dist == "Exponential" | Dist == "Normal" | Dist == "Uniform")){
-      stop("For logistic regression select the distribution 'Exponential', \n 'Normal' or 'Uniform' to generate the covarate data")
-    }
-  }
-
-  if(family == "poisson"){
-    if(!(Dist == "Normal" | Dist == "Uniform")){
-      stop("For poisson regression select the distribution 'Normal' \n or 'Uniform' to generate the covarate data")
-    }
-  }
-
-  if(family %in% "linear"){
-    if(Dist %in% "Normal"){
-      X<-replicate(No_Of_Var,stats::rnorm(n = N, mean = Dist_Par$Mean, sd = sqrt(Dist_Par$Variance)))
-    }
-    if(Dist %in% "Uniform"){
-      X<-replicate(No_Of_Var,stats::runif(n = N, min = Dist_Par$Min, max = Dist_Par$Max))
-    }
-    Complete_Data<-cbind(1,X);
-    colnames(Complete_Data)<-c(paste0("X",0:ncol(X)))
-    Residual<-stats::rnorm(n=N,mean=0,sd=sqrt(Dist_Par$Error_Variance))
-    Y <- Complete_Data%*%Beta + Residual
-
-    Complete_Data<-cbind(Y,Complete_Data)
-    colnames(Complete_Data)<-c("Y",paste0("X",0:ncol(X)))
-
-    stats::lm(Y~.-1,data=as.data.frame(Complete_Data))->Results
-    Beta_Estimate<-stats::coefficients(Results)
-    Var_Epsilon_Estimate<-mean((Y-stats::fitted.values(Results))^2)
-
-    Outputs<-list("Basic"=list("N"=N,"Beta"=Beta,
-                               "Beta_Estimates"=Beta_Estimate,
-                               "Variance_Epsilon_Estimates"=Var_Epsilon_Estimate,
-                               "Distribution"=Dist,
-                               "Distribution_Parameter"=Dist_Par,
-                               "No_Of_Variables"=No_Of_Var),
-                  "Complete_Data"=Complete_Data)
-
-    class(Outputs)<-c("A_L_OptimalSubsampling","linear")
-    return(Outputs)
-  }
-  if(family %in% "logistic"){
-    if(Dist %in% "Exponential"){
-      X<-replicate(No_Of_Var,stats::rexp(n = N, rate = Dist_Par$Rate))
-    }
-    if(Dist %in% "Normal"){
-      X<-replicate(No_Of_Var,stats::rnorm(n = N, mean = Dist_Par$Mean, sd = sqrt(Dist_Par$Variance)))
-    }
-    if(Dist %in% "Uniform"){
-      X<-replicate(No_Of_Var,stats::runif(n = N, min = Dist_Par$Min, max = Dist_Par$Max))
-    }
-
-    Complete_Data<-cbind(1,X)
-    colnames(Complete_Data)<-c(paste0("X",0:ncol(X)))
-
-    Linear_Predictor_Data <- Complete_Data%*%Beta
-    Pi_Data <- 1-1/(1+exp(Linear_Predictor_Data))
-    Y <- stats::rbinom(n=N,size=1,prob = Pi_Data)
-
-    Complete_Data<-cbind(Y,Complete_Data)
-    colnames(Complete_Data)<-c("Y",paste0("X",0:ncol(X)))
-
-    stats::glm(Y~.-1,data=as.data.frame(Complete_Data),family = "binomial")->Results
-    Beta_Estimate<-stats::coefficients(Results)
-
-    Outputs<-list("Basic"=list("N"=N,"Beta"=Beta,
-                               "Beta_Estimates"=Beta_Estimate,
-                               "Distribution"=Dist,
-                               "Distribution_Parameter"=Dist_Par,
-                               "No_Of_Variables"=No_Of_Var),
-                  "Complete_Data"=Complete_Data)
-
-    class(Outputs)<-c("A_L_OptimalSubsampling","logistic")
-    return(Outputs)
-  }
-  if(family %in% "poisson"){
-    if(Dist %in% "Normal"){
-      X<-replicate(No_Of_Var,stats::rnorm(n = N, mean = 0, sd = 1))
-    }
-    if(Dist %in% "Uniform"){
-      X<-replicate(No_Of_Var,stats::runif(n = N, min = 0, max = 1))
-    }
-
-    Complete_Data<-cbind(1,X)
-    colnames(Complete_Data)<-c(paste0("X",0:ncol(X)))
-
-    Linear_Predictor_Data <- Complete_Data%*%Beta
-    Lambda_Data <- exp(Linear_Predictor_Data)
-    Y <- stats::rpois(n=N,lambda = Lambda_Data)
-
-    Complete_Data<-cbind(Y,Complete_Data)
-    colnames(Complete_Data)<-c("Y",paste0("X",0:ncol(X)))
-
-    stats::glm(Y~.-1,data=as.data.frame(Complete_Data),family = "poisson")->Results
-    Beta_Estimate<-stats::coefficients(Results)
-
-    Outputs<-list("Basic"=list("N"=N,"Beta"=Beta,
-                               "Beta_Estimates"=Beta_Estimate,
-                               "Distribution"=Dist,
-                               "Distribution_Parameter"=Dist_Par,
-                               "No_Of_Variables"=No_Of_Var),
-                  "Complete_Data"=Complete_Data)
-
-    class(Outputs)<-c("A_L_OptimalSubsampling","poisson")
-    return(Outputs)
-  }
-}