7
$\begingroup$

I want to add two assumptions, so I can get this probability density function to equal 1, though I can't get a solution.

Integrate[(L r x^(r - 2))/(r - 1)! e^(-L x), {x, 0, Infinity}, 
 Assumptions -> r > 0, Assumptions -> L > 0]

My equation. With the assumptions r > 0 and L > 0

$\endgroup$
2
  • $\begingroup$ The exponential function is E^(...) or Exp[...]. $\endgroup$ Mar 12, 2013 at 12:12
  • $\begingroup$ Can only give one Assumptions-> option but it can have a List or conjunction (And). So Assumptions->{r>0,L>0} would be a correct syntax. I get Integrate[(L r x^(r - 2))/(r - 1)! e^(-L x), {x, 0, Infinity}, Assumptions ->{r>0,L>0}] // InputForm Out[2]//InputForm= ConditionalExpression[(L*r*Gamma[-1 + r]*(L*Log[e])^(1 - r))/(-1 + r)!, r > 1 && ((Re[Log[e]] == 0 && r < 2) || Re[Log[e]] > 0)] $\endgroup$ Mar 13, 2013 at 13:37

3 Answers 3

7
$\begingroup$

You can make multiple assumptions by passing a list:

Integrate[(L r x^(r - 2))/(r - 1)! e^(-L x), {x, 0, Infinity}, 
    Assumptions -> {r > 0, L > 0}]

Or, if you prefer, a boolean expression:

Integrate[(L r x^(r - 2))/(r - 1)! e^(-L x), {x, 0, Infinity}, 
    Assumptions -> r > 0 && L > 0]
$\endgroup$
1
  • $\begingroup$ Hmm ok i tried to get a solution, but as far as i can see, it doesnt not give you 1. Maybe someone have a solution? $\endgroup$
    – Anru Narenthirarajah
    Mar 13, 2013 at 11:49
1
$\begingroup$

You can combine both assumptions :

Integrate[(L r x^(r - 2))/(r - 1)! e^(-L x), {x, 0, Infinity}, 
  Assumptions -> {r > 0, L > 0}]
$\endgroup$
1
$\begingroup$

Your integral is available analytically with:

Integrate[(L*r *x^(r - 2))/(r - 1)! Exp[-L*x], {x, 0, Infinity}, Assumptions -> {r > 1, L > 0}]

... with the result

(L^(2 - r) r)/(-1 + r)

This is unity only for careful choices of L and r so I wonder if your question isn't about a normalization rather than an absolute check on a probability density function.

$\endgroup$
2
  • $\begingroup$ The assumptions are slightly different (r>1) and allow Mathematica to find the analytic solution. Also addresses the likely mistyping of 'e' for 'E'... $\endgroup$
    – SEngstrom
    Mar 19, 2013 at 16:37
  • $\begingroup$ Sorry, my mistake, you are right! $\endgroup$ Mar 19, 2013 at 17:26

Your Answer

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