试图通过单索引模型进行预测时出现两个R错误消息

Question

I am attempting to use the Single Index model to estimate alpha beta and sigma2_ei for 3 stocks and the TSX. Here is my current R code

Importing the Data

setwd("~/Desktop/R Data Sets") Ass2Data1 <- read.csv("~/Desktop/R Data     Sets/Ass2DataSheet1.csv") View(Ass2Data1)

Transforming the data into matrix form

b <- as.matrix(Ass2Data1)

Generating Initial Vectors and Matrices

x <- rep(0,60) 
xx <- matrix(x, ncol=4, nrow=3)
stock <- rep(0,3) 
alpha <- rep(0,3) 
beta <- rep(0,3) 
mse <- rep(0,3) 
Rbar <- rep(0,3) 
Ratio <- rep(0,3)

col1 <- rep(0,3) 
col2 <- rep(0,3) 
col3 <- rep(0,3) 
col4 <- rep(0,3) 
col5 <- rep(0,3)

Regressing each stock on the index and recording results

for(i in 1:3){
alpha[i] <- lm(data=Ass2Data1,formula=Ass2Data1[,1] ~ Ass2Data1[,4]$coefficients[1])
beta[i] <- lm(data=Ass2Data1,formula=Ass2Data1[,2] ~ Ass2Data1[,4]$coefficients[2])
Rbar[i] <- alpha[i]+beta[i]*mean(b[,4])
mse[i] <-sum(lm(data=Ass2Data1,formula=Ass2Data1[,i] ~ Ass2Data1[,4])$residuals2)/(nrow(b)-2)
Ratio[i] <- (Rbar[i]/beta[i])
stock[i] <- i }

Creating table

xx <- (cbind(stock,alpha,beta,Rbar,mse,Ratio))

However I keep getting the following error messages:

Error in Ass2Data1[, 4]$coefficients : $ operator is invalid for atomic vectors

and

Error in beta[i] * mean(b[, 4]) : non-numeric argument to binary operator

If anybody could point me in the right direction with respect to what I'm doing wrong here it would be greatly appreciated.

Answer 1

Check ?lm , change the closing )

alpha[i] <- lm(data=Ass2Data1,formula=Ass2Data1[,1] ~ Ass2Data1[,4])$coefficients[1]
beta[i] <- lm(data=Ass2Data1,formula=Ass2Data1[,2] ~ Ass2Data1[,4])$coefficients[2]

Better way is to get coefficients via coef :

alpha[i] <- coef(lm(data=Ass2Data1,formula=Ass2Data1[,1] ~ Ass2Data1[,4]))[1]
 beta[i] <- coef(lm(data=Ass2Data1,formula=Ass2Data1[,2] ~ Ass2Data1[,4]))[2]