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
  • 1
    $\begingroup$ You could also use InverseErf, but that's cumbersome too. $\endgroup$
    – user484
    Jun 12, 2015 at 4:32

1 Answer 1

7
$\begingroup$

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

InverseCDF[NormalDistribution[mu, sigma], prob]

$\endgroup$
3
  • 3
    $\begingroup$ Quantile[NormalDistribution[mu, sigma], prob] saves 2 ;-) $\endgroup$
    – ciao
    Jun 12, 2015 at 4:48
  • $\begingroup$ @ciao +1 for teaching me about a new function :D $\endgroup$
    – 1110101001
    Jun 12, 2015 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, 2015 at 5:15

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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