2
$\begingroup$

In R and TI-89 you can do

qnorm(prob, mu, sigma) or invNorm(prob, mu, sigma) to get a value of x such that $P(Z < x)$ along a normal curve defined by the specified mean and standard deviation equals prob.

While I could do something like:

Solve[CDF[NormalDistribution[mu, sigma], x] == prob, x]

this is cumbersome.

While I could also use Mathematica's RLink functionality, I would like to know whether there is a native command that does this.

$\endgroup$
  • 1
    $\begingroup$ You could also use InverseErf, but that's cumbersome too. $\endgroup$ – Rahul Jun 12 '15 at 4:32
6
$\begingroup$

I think this is the most compact I can get without defining my own function:

InverseCDF[NormalDistribution[mu, sigma], prob]

$\endgroup$
  • 3
    $\begingroup$ Quantile[NormalDistribution[mu, sigma], prob] saves 2 ;-) $\endgroup$ – ciao Jun 12 '15 at 4:48
  • $\begingroup$ @ciao +1 for teaching me about a new function :D $\endgroup$ – 1110101001 Jun 12 '15 at 4:54
  • 2
    $\begingroup$ But why not define q[prob_, mu_, sigma_] := InverseCDF[...]? You'd be using the native command and saving keystrokes. $\endgroup$ – ilian Jun 12 '15 at 5:15

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.