0

I am running a program where I conduct an OLS regression and then I subtract the coefficients from the actual observations to keep the residuals.

model1 = lm(data = final, obs ~ day + poly(temp,2) + prpn + school + lag1) # linear model  
predfit = predict(model1, final) # predicted values

residuals = data.frame(final$obs - predfit) # obtain residuals

I want to bootstrap my model and then do the same with the bootstrapped coefficients. I try doing this the following way:

lboot <- lm.boot(model1, R = 1000)
predfit = predict(lboot, final)

residuals = data.frame(final$obs - predfit) # obtain residuals

However, that does not work. I also try:

boot_predict(model1, final,  R = 1000, condense = T, comparison = "difference")

and that also does not work.

How can I bootstrap my model and then predict based of that?

StupidWolf
  • 45,075
  • 17
  • 40
  • 72
Santi
  • 1
  • 1

1 Answers1

1

If you're trying to fit the best OLS using bootstrap, I'd use the caret package. 

library(caret)

#separate indep and dep variables
indepVars = final[,-final$obs]
depVar = final$obs

#train model
ols.train = train(indepVars, depVar, method='lm',
                  trControl = trainControl(method='boot', number=1000))

#make prediction and get residuals
ols.pred = predict(ols.train, indepVars)
residuals = ols.pred - final$obs
Todd Burus
  • 963
  • 1
  • 6
  • 20