how do you calculate CI in R and explain the method?

Asked Jul 08 '16 at 13:45

Active Jul 08 '16 at 14:13

Viewed 24 times

I am trying to perform prediction using R with confidence interval:

mod <- glm(y ~ x, data = foo, family = binomial)
preddata <- with(foo, data.frame(x = seq(min(x), max(x), length = 100))
preds <- predict(mod, newdata = preddata, type = "link", se.fit = TRUE)

critval <- 1.96 
upr <- preds$fit + (critval * preds$se.fit)
lwr <- preds$fit - (critval * preds$se.fit)
fit <- preds$fit

I am trying to understand why are we multiplying preds$sef.fit by 1.96? Can somebody explain to me the importance of 1.96, please?

asked Jul 08 '16 at 13:45

user1471980

@ZheyuanLi, I appreciate your answer but what is qnorm(0.025)? – user1471980 Jul 08 '16 at 13:50
3

[This](https://en.wikipedia.org/wiki/1.96) will clarify it for you. – Sotos Jul 08 '16 at 13:50
1.96 is a famous number in introductory statistics. type "1.96 statistics" (without quotes) into a search engine, and you will get bunches of hits. – lmo Jul 08 '16 at 13:51
2

wow, @Sotos, I didn't know that 1.96 had its own Wikipedia page ... – Ben Bolker Jul 08 '16 at 13:59
@BenBolker, to be honest neither did I. I just stumbled upon it now. :) – Sotos Jul 08 '16 at 14:01

how do you calculate CI in R and explain the method?

0 Answers0