Solve this problem that involves a probability density function

Click For Summary
The discussion revolves around solving a problem involving a probability density function (PDF) defined as f(u) = 3u - (3u^2)/(2k) with limits 0 ≤ u ≤ k. The value of k is determined to be 2/3, and calculations for the expected value E(T) and variance Var(T) are presented, with some corrections made along the way. Notation issues are raised, particularly regarding the distinction between PDF and cumulative distribution function (CDF), with suggestions for standard notation. Participants confirm the calculations and express their intent to complete the remaining parts of the problem. The conversation emphasizes clarity in notation and accuracy in statistical calculations.
chwala
Gold Member
Messages
2,827
Reaction score
415
Homework Statement
see attached
Relevant Equations
understanding of continous and discrete distribution
I am refreshing on this; ..after a long time...

Note that i do not have the solution to this problem.

1667393644541.png
I will start with part (a).

##f(u)= 3u-\dfrac{3u^2}{2k}## with limits ##0≤u≤k##

it follows that,

##3k - \dfrac{3k}{2}=1##

##\dfrac{3k}{2}=1##

##k=\dfrac {2}{3}##

For part (b),

##E(T)=\int_0^{\frac{2}{3}} u⋅(3-\dfrac{9}{2}u )du=\left[\dfrac{3}{2}×\dfrac{4}{9}-\dfrac{3}{2}×\dfrac{8}{27}\right]=\dfrac{6-4}{9}= \dfrac{2}{9}##Ok let me know if that's correct...
 
Last edited:
Physics news on Phys.org
Those both look OK to me. Do you know how to do the last two parts?

One minor  issue
chwala said:
##f(u)= 3u-\dfrac{3u^2}{2k}## with limits ##0≤u≤k##
This is bad notation, since ##f## it's already defined to be the pdf, not the cdf.
 
Office_Shredder said:
Those both look OK to me. Do you know how to do the last two parts?

One minor  issue

This is bad notation, since ##f## it's already defined to be the pdf, not the cdf.
I will work on the last parts too...refreshing on this...should be fine ...Will post once done.
 
Office_Shredder said:
Those both look OK to me. Do you know how to do the last two parts?

One minor  issue

This is bad notation, since ##f## it's already defined to be the pdf, not the cdf.
I thought pdf is generally defined to be ##f_{u}## being the derivative of the continuous distribution ##f(u)## ...different books have different notations/language that I pretty find to be time waster...my focus is on the concept...

Which is the right notation? Thanks @Office_Shredder
 
Last edited:
chwala said:
Which is the right notation? Thanks @Office_Shredder
In my stats class we used ##f_{X}(x)## for the PDF and ##F_{X}(x)## for the CDF for a given continuous random variable ##X##. I think that this is fairly standard, because most books and lecture notes I found online used it.
 
Office_Shredder said:
Those both look OK to me. Do you know how to do the last two parts?

One minor  issue

This is bad notation, since ##f## it's already defined to be the pdf, not the cdf.
##E(T^2)=\int_0^{\frac{2}{3}} u^2⋅(3-\dfrac{9}{2}u )du##

##=\int_0^{\frac{2}{3}} (3u^2-\dfrac{9u^3}{2} )du=\left[u^3-\dfrac{9u^4}{8} \right]##

##=\left[\dfrac{8}{27}-\dfrac{9}{8}×\dfrac{16}{81}\right]=\dfrac{8-6}{27}= \dfrac{2}{27}##

##⇒Var (T) = \dfrac{24-4}{81}=\dfrac{20}{81}##

...this part is easy and straightforward. Bingo!
 
Where does the 24 in the variance come from? Shouldn't it be a 6?
 
Office_Shredder said:
Where does the 24 in the variance come from? Shouldn't it be a 6?
True...let me amend that.
 
chwala said:
##E(T^2)=\int_0^{\frac{2}{3}} u^2⋅(3-\dfrac{9}{2}u )du##

##=\int_0^{\frac{2}{3}} (3u^2-\dfrac{9u^3}{2} )du=\left[u^3-\dfrac{9u^4}{8} \right]##

##=\left[\dfrac{8}{27}-\dfrac{9}{8}×\dfrac{16}{81}\right]=\dfrac{8-6}{27}= \dfrac{2}{27}##

##⇒Var (T) = \dfrac{6-4}{81}=\dfrac{2}{81}##

...this part is easy and straightforward. Bingo!