为R中的16组参数生成10000个样本的有效方法是什么

我被要求解决以下问题:

但是我很难用16组参数生成10000个样本.这就是我所做的:

library(MASS)
library(data.table)
library(Matrix)
library(stats)
set.seed(2000)
options(scipen=100)

N <- c(100, 500)
K <- c(1, 10)
ETA <- c(0.05, 1)
RHO <- c(0, 0.5)
reps <- 10000

DGP<-function(n,k,eta,rho){
  # generate D, then get \Lambda=DD', where z \sim N(0,\Lambda)
  D <- matrix(0, nrow = k, ncol = k)
  for (x in 1:k) {
    for (y in 1:k) {
      D[x, y] <- ifelse(x > y, 1, ifelse(x == y, 1, 0))
    }
  }
  # generate Z
  Z <- mvrnorm(n, mu = rep(0, k), Sigma = D%*%t(D))
  # generate error terms \epsilon, u, where y=\epsilon (since \beta-9) 
  # and u is the error term for x=z'\pi+u
  err <- mvrnorm(n, mu = rep(0, 2), Sigma = matrix(c(1, rho, rho, 1), ncol = 2))
  Y <- as.matrix(err[,1])
  u <- err[,2]
  # then we can get X
  X <- Z %*% rep(eta,k) + u
  return(data.frame(Y,X,Z))
}

# test if the DGP function works
DGP(500,1,1,0.5)

# generate the 16 samples
data<-list()

# I know this is inefficient but I gave up
para<-matrix(data=c(100,1,0.05,0,500,1,0.05,0,
     100,10,0.05,0,500,10,0.05,0,
     100,1,1,0,500,1,1,0,
     100,1,0.05,0.5,500,1,0.05,0.5,
     100,10,1,0,500,10,1,0,
     100,10,1,0.5,500,10,1,0.5,
     100,1,1,0.5,500,1,1,0.5,
     100,10,0.05,0.5,500,10,0.05,0.5),nrow=16,ncol=4,byrow=TRUE)


datalist<-function(){
  for (i in 1:16){
    df<-DGP(para[i,][1],para[i,][2],para[i,][3],para[i,][4])
    data[[i]]<-df
  }
  return(data)
}

sample<-datalist() # at least I have generated one sample

有没有办法更有效地为16组参数生成和存储10000个样本？