CODEWITHSUNDEEP
r
yefersonG
yefersonG

Reputation: 153

confidence interval for the average quantity

Hello I would like to know what is the command to do the following problem:

Determine a 95% confidence interval for the average number of winning games. swim, limited 2 000 in Rstudio

viex<-c(2205,2096, 1847, 1903, 1457, 1848, 1564, 1821, 2577, 2476, 1984, 1917, 1761, 1709, 1901, 2288, 2072, 2861, 2411, 2289, 2203 ,2592, 2053 ,1979, 2048, 1786, 2876, 2560)

clasy<-c(10, 11, 11, 13, 10, 11, 10 ,11,  4 , 2,  7, 10,  9,  9,  6,  5,  5,  5,  6,  4 , 3 , 3 , 4, 10,  6 , 8,  2 , 0)

modelo = lm(clasy~ viex)
modelo

that's what i've done

Upvotes: 1

Views: 51

Answers (1)

Rui Barradas
Rui Barradas

Reputation: 76663

This comes directly from the definition of confidence interval.

ci_regression <- function(x, conf = 0.95){
  est <- coef(summary(x))[, 1]
  se <- coef(summary(x))[, 2]
  qq <- qt(1 - (1 - conf)/2, df = x$df.residual)
  cbind(lower = est - qq*se, upper = est + qq*se)
}

ci_regression(modelo)
#                   lower        upper
#(Intercept) 16.246064040 27.330437725
#viex        -0.009614347 -0.004435854

In the case of linear regression, there is a confint.lm method for objects of class "lm".
Simply run

confint(modelo)
#                   2.5 %       97.5 %
#(Intercept) 16.246064040 27.330437725
#viex        -0.009614347 -0.004435854

Upvotes: 1

Related Questions