[Rd] Getting high precision values from qnorm in the tail

2017-04-16 Thread Sheldon Maze
Hello All I am looking for high precision values for the normal distribution in the tail,(1e-10 and 1 - 1e-10) as the R package that I am using sets any number which is out of this range to these values and then calls the qnorm and qt function. What I have noticed is that the qnorm implementation

Re: [Rd] Getting high precision values from qnorm in the tail

2017-04-16 Thread Spencer Graves
rtfm: help('qnorm') identifies arguments you overlooked. 1-x generates roundoff error. Try the following: x <- 10^(-(1:10)) qx <- qnorm(x) q1x <- qnorm(1-x) qlx <- qnorm(x, lower=FALSE) > cbind(x, qx, q1x, qlx, qx.1x=qx+q1x, qx.lx=qx+qlx) xqx q1x qlx qx.1