Anyone available for Stat Homework that's due tomorrow? Gamma distribution - pasting the image of the problem in the comments. Thank you in advance!!
$$ \Large E(X) = \int x \cdot f(x) $$
is that the same though, as \[E(\frac{ 1 }{ X })\]?
and what is x? f(x) is my gamma function, correct? Is x=6?
almost the same
$$ \Large E(\frac1X) = \int\frac1 x \cdot f(x) $$
$$ \Large E(\frac1X) = \int_{-\infty }^{\infty} ~ \frac1 x \cdot \frac{1}{\Gamma (6)\cdot 2^6 }\cdot x^{6-1}e^{\frac{-x}{2}} $$
HI not to butt in, but i think there might be a way to do this directly also, isn't the integral from 0 to \(\infty\)? gamma not defined for \(x<0\) if i am not mistaken
check the formula for the kth moment on page 36 here http://ocw.mit.edu/courses/mathematics/18-443-statistics-for-applications-fall-2006/lecture-notes/lecture6.pdf
looks like it might be t \[\frac{\Gamma(5)}{\Gamma(6)\frac{\pi}{2}^{-1}}\]
\[\Gamma(5)=4!, \Gamma(6)=5! \] so it might be just \[\frac{\pi}{5}\]
nope bad algebra, \[\frac{\pi}{10}\]
@perl look reasonable or am i way off base?
oh another mistake, i completely ignored the \(2^6\) in the denominator
right, start at x = 0
$$ \Large E(\frac1X) = \int_{0 }^{\infty} ~ \frac1 x \cdot \frac{1}{\Gamma (6)\cdot 2^6 }\cdot x^{6-1}e^{\frac{-x}{2}} $$
lol i put a \(\pi\) there instead of a 2
it is one
\[\gamma(6)=5!, 5!\times 2^6=7680\]
\Gamma
yeah that one
so wolfram says E(1/x) = .10 http://www.wolframalpha.com/input/?i=int%281%2Fx*x^5*exp%28-%281%2F2%29*x%29%2F%28gamma%286%29*2^6%29%2C+x+%3D+0+..+infinity%29
\[\frac{1}{7680}\int_0^{\infty}x^4e^-{\frac{x}{2}}\]
damn i can't see what i am typing, lotsa typos
\[\frac{1}{7680}\int_0^{\infty}x^4e^{\frac{x}{2}}dx\]
aarrgghh
what about the (1/X) being multiplied in that integral?
\[\frac{1}{7680}\int_0^{\infty}x^4e^{-\frac{x}{2}}dx\]
that turns \(x^5\) in to \(x^4\)
oh, ok I see that
you can do integration by parts a few times
lol yeah quite a few
bet we can do it quicker
\[\frac{\Gamma(5)}{\Gamma(6)2^{-1}}\]
no that is not working dang i wonder why...
ooh because i am an idiot!!
\[\frac{\Gamma(5)}{\Gamma(6)(\frac{1}{2})^{-1}}\]
\[\frac{4!}{2\times 5!}\]
yayyy!!!! wasn't that easy?
haha! Thank you to both of you for jumping in to help me!! I am totally with you until the end @misty1212 - I have no idea where that last equation came from. :)
ohhh...we haven't covered moments yet. Guess that's my problem. :(
damn it \[E(X^k)=\frac{\Gamma(\alpha+k)}{\Gamma(\alpha)\beta^k}\]
in your case \(\alpha=6, \beta=\frac{1}{2},k=-1\)
and [\Gamma(\alpha)=(\alpha-1)!\] for integer \(\alpha\)
is that formula given in the pdf
\[\Gamma(\alpha)=(\alpha-1)!\]
damn i hate not having the previewq
sure is a lot easier than integrating by parts 4 times !!!
yeah that is the formula in the pdf, take a look it is derived there
with that formula takes like two seconds if you do it correctly and not mess about like i did 4 or 5 times
You guys are awesome! Thank you so much!
yes thats pretty nifty :)
$$ \Large { \mathbb{E}(X^k) = \frac{\Gamma(\alpha+k)}{\Gamma(\alpha)\beta^k} } $$
oh by the way, lowercase \gamma is used for euler's constant. thats why i was getting a strange answer on maple
I think I need to show my homework with the integration by parts, however I've done this at least 4 times and never get the 1/10 that Wolfram shows. I keep getting -4
you can expedite the integration by parts by making a table
@misty1212 I'm surprised this formula is not mentioned in the wikipedia article. Unless i missed it http://en.wikipedia.org/wiki/Gamma_distribution
@BrighterDays want to make a table?
I don't know about making tables?? :)
|dw:1428553530664:dw|
Join our real-time social learning platform and learn together with your friends!