how can i integral a function which was written in matlab but there is wrong , i dont no where it is :
syms tau
syms y
f= @(y,tau)y*exp(-0.5.*y.^2/ tau)
expecty=integral2(f, -2,4,0,Inf,'auto')

 Risposta accettata

Birdman
Birdman il 16 Mar 2018

1 voto

integral2 is defined for function handle, therefore you do not need symbolic variables:
f= @(y,tau)y.*exp(-0.5.*y.^2./ tau)
expecty=integral2(f, -2,4,0,Inf,'auto')

6 Commenti

if i use it as a defined , the result is:
expecty =
NaN
Torsten
Torsten il 16 Mar 2018
The result is correct.
Did you expect something finite for the integral ?
elham kreem
elham kreem il 16 Mar 2018
do you mean that :
expecty = nan
is correct?
Steven Lord
Steven Lord il 16 Mar 2018
When y = 0 and tau = 0, you're computing 0.*exp(0./0). 0./0 is NaN, exp(NaN) is NaN, and 0*NaN is NaN. Once NaN enters a calculation, it tends to propagate throughout that calculation.
Torsten
Torsten il 16 Mar 2018
Neglecting numerical aspects like division by 0, your integral is +Inf.
Best wishes
Torsten.
gentelmen , thanks for your answers , but there is something wrong?
because the result is not convincing .
if i integrate the function with syms y and tau , are the code right and the result?
syms tau
syms y
fun = @(y,tau) (1/tau) * y * exp((-0.5*y^2 / tau)
f1 = int(fun , y ,tau )
is this mean double integral , for y and tau?
the result is
exp(-y^2/(2*tau)) - exp(-tau/2)
is the result convincing?

Accedi per commentare.

Più risposte (0)

Tag

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by