90% or 95% confidence intervals calculated by predict.lm

Asked Nov 21 '19 at 22:23

Active Nov 21 '19 at 22:23

Viewed 42 times

I am fitting a model by lm(), predict data wich gives me also confidence intervals (lwr, upr), and plot a scatterplot with regression line with confidence intervals. Yes, it works nicely.

My only question - I could not find it in the documentation or elsewhere specifically for this function - is: are these calculated confidence intervals (presumably) 95% (or 90%?) confidence intervals, or can one change that?

df    <- data.frame(x = x, y = y)
fit   <- lm(y ~ x, data = df))
new.x <- seq(min(df$x), max(df$x), length.out=100)
pred  <- predict(fit, newdata = data.frame(x=new.x), 
                 interval = 'confidence')

asked Nov 21 '19 at 22:23

Martin

5

See `?predict.lm` - there is a `level=` argument which is set to `0.95` by default – thelatemail Nov 21 '19 at 22:26
@thelatemail: Uhps, I overlooked that line. +1, would accept as answer. – Martin Nov 21 '19 at 22:33

90% or 95% confidence intervals calculated by predict.lm

0 Answers0