The Student Room Group

Intergration by parts

Attached I have a picture of an integration I am struggling with.

I am trying to use integration by parts, making u equal to "e to the power of minus x squared" and dv/dx equal to x but I keep getting the former (e to the power of minus x squared) in the integral which I can't/don't know how to integrate.

I can't seem to find any other way around it.

Please help.

Thank you.
Reply 1
Original post by Nishap
Attached I have a picture of an integration I am struggling with.

I am trying to use integration by parts, making u equal to "e to the power of minus x squared" and dv/dx equal to x but I keep getting the former (e to the power of minus x squared) in the integral which I can't/don't know how to integrate.

I can't seem to find any other way around it.

Please help.

Thank you.


Try a substitution: u=x2 u = x^{2}
(edited 10 years ago)
Reply 2
Original post by Blazy
Try a substitution: u=x2 u = x^{2}


but then what would dv/dx equal if the substitution was x squared?
Reply 3
Original post by Nishap
Attached I have a picture of an integration I am struggling with.

I am trying to use integration by parts, making u equal to "e to the power of minus x squared" and dv/dx equal to x but I keep getting the former (e to the power of minus x squared) in the integral which I can't/don't know how to integrate.

I can't seem to find any other way around it.

Please help.

Thank you.


Can you write out explicitly what you get from integrating by parts?
Reply 4
Original post by Blazy
Try a substitution: u=x2 u = x^{2}


Ah sorry! Do you mean integration by substitution rather than by parts?
Reply 5
By using integration by substitution (substituition u= minus x squared), I've got the answer of 0.5 (i.e. a half). Hoping this is correct now.
Reply 6
you don`t need integration by parts.

If you make a slight alteration, you end up with an integral of the form:

0f(x)ef(x)dx\int_{0}^{\infty} f ' (x) e^{f(x)}dx

you can alter it thus:

....12×limϵ>0ϵ(2x)ex2dx\displaystyle .... \equiv \frac{-1}{2}\times lim_{\epsilon->\infty^{-}} \int_{0}^{\epsilon} (-2x) e^{-x^2}dx
Reply 7
Original post by Nishap
By using integration by substitution (substituition u= minus x squared), I've got the answer of 0.5 (i.e. a half). Hoping this is correct now.


Yes, that's correct.
Reply 8
Use polar coordinates.
Reply 9
Original post by Arieisit
Use polar coordinates.


You don't need polar coordinates for this integral - it's a simple substitution that's required

(you may be thinking of the normal probability integral which is easiest to do by squaring and transforming to polar coordinates)
Reply 10
Original post by davros
You don't need polar coordinates for this integral - it's a simple substitution that's required

(you may be thinking of the normal probability integral which is easiest to do by squaring and transforming to polar coordinates)


Yea, I forgot there was the x otherwise

I=ex2dx\displaystyle I=\int_{- \infty}^{\infty} e^{-x^2} dx

I2=(ex2dx)2=(ex2dx)(ex2dx)\displaystyle I^2= (\int_{- \infty}^{\infty} e^{-x^2} dx)^2=(\int_{- \infty}^{\infty} e^{-x^2} dx)(\int_{- \infty}^{\infty} e^{-x^2} dx)

I2=(ex2dx)(ey2dy)\displaystyle I^2= (\int_{- \infty}^{\infty} e^{-x^2} dx)(\int_{- \infty}^{\infty} e^{-y^2} dy)

I2=e(x2+y2)dxdy\displaystyle I^2= \int_{- \infty}^{\infty}\int_{- \infty}^{\infty} e^{-(x^2+y^2)} dxdy Switching to polar coordinates r=x2+y2r=x^2+y^2

Unparseable latex formula:

\displaystyle I^2= \int_{0}^{2 \pi} \int_{0}^{ \infty} re^{-r^2} dr \d \theta= -\dfrac{1}{2} \int_{0}^{2 \pi} \left[ re^{-r^2} \right]_0^{\infty} d \theta =\int_{0}^{2 \pi} -1 d \theta= \pi



I2=πI=π\therefore I^2= \pi \Rightarrow I=\sqrt{\pi}.



Posted from TSR Mobile
Reply 11
Original post by Arieisit
Yea, I forgot there was the x otherwise

I=ex2dx\displaystyle I=\int_{- \infty}^{\infty} e^{-x^2} dx

I2=(ex2dx)2=(ex2dx)(ex2dx)\displaystyle I^2= (\int_{- \infty}^{\infty} e^{-x^2} dx)^2=(\int_{- \infty}^{\infty} e^{-x^2} dx)(\int_{- \infty}^{\infty} e^{-x^2} dx)

I2=(ex2dx)(ey2dy)\displaystyle I^2= (\int_{- \infty}^{\infty} e^{-x^2} dx)(\int_{- \infty}^{\infty} e^{-y^2} dy)

I2=e(x2+y2)dxdy\displaystyle I^2= \int_{- \infty}^{\infty}\int_{- \infty}^{\infty} e^{-(x^2+y^2)} dxdy Switching to polar coordinates r=x2+y2r=x^2+y^2

Unparseable latex formula:

\displaystyle I^2= \int_{0}^{2 \pi} \int_{0}^{ \infty} re^{-r^2} dr \d \theta= -\dfrac{1}{2} \int_{0}^{2 \pi} \left[ re^{-r^2} \right]_0^{\infty} d \theta =\int_{0}^{2 \pi} -1 d \theta= \pi



I2=πI=π\therefore I^2= \pi \Rightarrow I=\sqrt{\pi}.



Posted from TSR Mobile


One of my favorite integrals - shame it wasn't needed here!

Quick Reply

Latest