Estimating Coefficients of Multiple Regression

Question

I have the following problem:

I'm using the data "Railtrail" from this library "mosaicData". I already have the coeffiecient of this following linear regression model: lm(volume ~ hightemp + cloudcover + weekday, data = RailTrail) , compute for the population.

Now, I need to estimate the coeffiecient of that model with samples and to build a confidence interval (95%). So I need to compute all the coefficients of the data samples previously generated. I was asked to use a loop 'for' but I don't know how to compute the LR models. I also need to store the coefficient obtained.

I tried to do it doing this

trial <- list()
set.seed(101)

for(i in 1:100){
  trial[[i]] <- RailTrail %>% 
    lm(volume ~ hightemp + cloudcover + weekday, data = RailTrail)
}

but I get the following error:

Error in xj[i]: invalid subscript type 'language'

Thank you,

Don't hesitate to ask further precision if my request is not clear.

Francisco

Answer 1

Do you mean the confidence interval for the model parameters? If so, this example I hope illustrates a succinct way of doing so:

model <- lm(mpg ~ cyl + disp + gear, data = mtcars)
bind_cols(broom::tidy(model), broom::confint_tidy(model))