library(glmnet)

as <- read.csv("E://User tgvf554//October and November//Tian Sun//LASSO regression.csv")
head(as)
x<- model.matrix(X1~.,as)
y <- as$X1

cvRs <- cv.glmnet(x,y,
                  alpha=1,
                  nfolds=10,
                  family="gaussian",
)

plot(cvRs)

lambda_min <- cvRs$lambda.min
lambda_min

#best value of lambda
lambda_1se <- cvRs$lambda.1se
lambda_1se
coef(cvRs,s=lambda_min)
Active.Index<-which(coefficients!=0)
Active.coefficients <- coefficients(Active.Index)
####Lasso Process diagram for filtering dynamic variables 
fit <- glmnet(x,y,family = "gaussian")

plot(fit,xvar="lambda",label = "lambda")


coefficients <-coef(fit,s=cvRs$lambda.min)

coefficients