# Need help with calculus problem

1. Dec 6, 2004

can anyone figure this out? i keep getting undefined somewhere but my teacher says there is a correct answer.

heres the problem.

.0065 times the integral of e to the (-.0139(t-48)squared) power.

i cant find the integral cause i get undefined when the chain rule is used to chain out the exponent.

2. Dec 6, 2004

### fourier jr

i don't think there is a correct answer. not with "elementary functions" anyway.

3. Dec 7, 2004

### Galileo

$$\exp\left(-a(t-t_0)^2\right)$$ does not have an elementary antiderivative.
You can get an exact expression of the integral though if the limits of integration extend to infinity.

What are the limits of integration in your problem?

4. Dec 7, 2004

### harsh

Try graphing the equation. I believe the gaussian (bell - curve) is also this kind, for which the answer should then be 1. Not sure though.

5. Dec 7, 2004

### Zurtex

Presumably you're looking for a numerical solution right? Otherwise the solution is the error function related which is not elementary.

6. Dec 7, 2004

well the thing is this is a problem in our calculus book in the section under logarithmic and exponential integration. which is wierd. there is nothing else to the problem except what i have given here. the book asks for the answer to that probability equation and wants it as a percentage. while its great that i have the answer, i dont know how to get it because i get undefined

7. Dec 8, 2004

### HallsofIvy

Staff Emeritus
It might be a good idea to tell us what the problem actually is! The way you posed the problem originally, it was an indefinite integral but it appears that the problem gave specific limits.

8. Dec 8, 2004

i thought i gave the restrictions? it was from 48-60. that is why i was saying i get undefined from the chain rule. sorry for the confusion

9. Dec 8, 2004

### Sick0Fant

10. Dec 9, 2004

### Galileo

Got that too: 0.04664079814

11. Dec 9, 2004