如何在 R 中为多重回归运行蒙特卡洛模拟？

如何解决如何在 R 中为多重回归运行蒙特卡洛模拟？

我想对预测 mpg 的多元回归模型运行蒙特卡罗模拟，然后评估每辆车比另一辆车具有更好性能（更低 mpg）的次数。这是我目前得到的

II I tI thI thiI thinI thinkI think I think @I think @gI think @guI think @guaI think @guacI think @guacaI think @guacamI think @guacamoI think @guacamolI think @guacamoleI think @guacamole 
I think @guacamole iI think @guacamole isI think @guacamole is I think @guacamole is gI think @guacamole is goI think @guacamole is gooI think @guacamole is goodI think @guacamole is good I think @guacamole is good fI think @guacamole is good foI think @guacamole is good forI think @guacamole is good for I think @guacamole is good for hI think @guacamole is good for heI think @guacamole is good for heaI think @guacamole is good for healI think @guacamole is good for healtI think @guacamole is good for healthI think @guacamole is good for health

此代码不会为 library(pacman) pacman::p_load(data.table,fixest,stargazer,dplyr,magrittr) df <- mtcars fit <- lm(mpg~cyl + hp,data = df) fit$coefficients[1] beta_0 = fit$coefficients[1] # Intercept beta_1 = fit$coefficients[2] # Slope (cyl) beta_2 = fit$coefficients[3] # slope (hp) set.seed(1) # Seed n = 1000 # Sample size M = 500 # Number of experiments/iterations ## Storage slope_DT <- rep(0,M) slope_DT_2 <- rep(0,M) intercept_DT <- rep(0,M) ## Begin Monte Carlo for (i in 1:M){ # M is the number of iterations # Generate data U_i = rnorm(n,mean = 0,sd = 2) # Error X_i = rnorm(n,mean = 5,sd = 5) # Independent variable Y_i = beta_0 + beta_1*X_i + beta_2*X_i +U_i # Dependent variable # Formulate data.table data_i = data.table(Y = Y_i,X = X_i) # Run regressions ols_i <- fixest::feols(data = data_i,Y ~ X) # Extract slope coefficient and save slope_DT_2[i] <- ols_i$coefficients[3] slope_DT[i] <- ols_i$coefficients[2] intercept_DT[i] <- ols_i$coefficients[1] } # Summary statistics estimates_DT <- data.table(beta_2 = slope_DT_2,beta_1 = slope_DT,beta_0 = intercept_DT) 创建任何系数我想知道如何将系数添加到模型中，然后预测结果并测试一辆汽车的 mpg 低于另一辆汽车的次数。例如，马自达 RX4 的预测 mpg 比 Datsun 710 低多少次。 关于如何使这项工作的一些想法？ 谢谢

解决方法

就像我在评论中指出的那样，您应该使用两个自变量。此外，我想向您推荐 lapply 函数，它使代码更短，因为您不需要初始化/存储部分。

estimates_DT <- do.call("rbind",lapply(1:M,function(i) {
  # Generate data
  U_i = rnorm(n,mean = 0,sd = 2) # Error
  X_i_1 = rnorm(n,mean = 5,sd = 5) # First independent variable
  X_i_2 = rnorm(n,sd = 5) #Second ndependent variable
  Y_i = beta_0 + beta_1*X_i_1 + beta_2*X_i_2 + U_i  # Dependent variable

  # Formulate data.table
  data_i = data.table(Y = Y_i,X1 = X_i_1,X2 = X_i_2)
  
  # Run regressions
  ols_i <- fixest::feols(data = data_i,Y ~ X1 + X2)  
  ols_i$coefficients
}))

estimates_DT <- setNames(data.table(estimates_DT),c("beta_0","beta_1","beta_2"))

要比较两辆车的预测，请定义以下函数，将要进行比较的两个汽车名称作为参数：

compareCarEstimations <- function(carname1="Mazda RX4",carname2="Datsun 710") {
  car1data <- mtcars[rownames(mtcars) == carname1,c("cyl","hp")]
  car2data <- mtcars[rownames(mtcars) == carname2,"hp")]
  
  predsCar1 <- estimates_DT[["beta_0"]] + car1data$cyl*estimates_DT[["beta_1"]]+car1data$hp*estimates_DT[["beta_2"]]
  predsCar2 <- estimates_DT[["beta_0"]] + car2data$cyl*estimates_DT[["beta_1"]]+car2data$hp*estimates_DT[["beta_2"]]
  
  list(
    car1LowerCar2 = sum(predsCar1 < predsCar2),car2LowerCar1 = sum(predsCar1 >= predsCar2)
  )
}

确保作为参数提供的名称是有效名称，例如在rownames(mtcars)。