CODEWITHSUNDEEP
r
Yihe Wang
Yihe Wang

Reputation: 59

Is there a R function to find Confident Interval given a value in R?

for example, if I have a dataset with X and Y, I know that I can find the confident interval of the linear model by confint(lm(Y~X)) But I am wondering if there is a way for me to find the CI for Y given that X = somevalue?

Upvotes: 0

Views: 28

Answers (1)

hxalchemy
hxalchemy

Reputation: 374

here is a solution

note: replace confidence with prediction if you want prediction interval


x <- 1:10
y <- 1:10 + 5 + runif(10)

model <- lm(y~x)


predict(model, newdata = data.frame(x = 1.5), interval = "confidence")

output>
fit      lwr      upr
1 7.164623 6.746502 7.582743

or multiple datapoints

predict(model, newdata = data.frame(x = c(1.5,2.5,10)), interval = "confidence")

output>
fit       lwr       upr
1  7.164623  6.746502  7.582743
2  8.139339  7.786692  8.491986
3 15.449707 14.996426 15.902989

or fitted values

predict(model,interval = "confidence")

output>
fit       lwr       upr
1   6.677265  6.223983  7.130546
2   7.651981  7.267546  8.036415
3   8.626696  8.303379  8.950014
4   9.601412  9.326281  9.876544
5  10.576128 10.328583 10.823674
6  11.550844 11.303298 11.798389
7  12.525560 12.250428 12.800691
8  13.500276 13.176958 13.823593
9  14.474991 14.090557 14.859426
10 15.449707 14.996426 15.902989

Upvotes: 2

Related Questions