• We are pleased to announce that the winner of our Feedback Prize Draw for the Winter 2024-25 session and winning £150 of gift vouchers is Zhao Liang Tay. Congratulations to Zhao Liang. If you fancy winning £150 worth of gift vouchers (from a major UK store) for the Summer 2025 exam sitting for just a few minutes of your time throughout the session, please see our website at https://www.acted.co.uk/further-info.html?pat=feedback#feedback-prize for more information on how you can make sure your name is included in the draw at the end of the session.
  • Please be advised that the SP1, SP5 and SP7 X1 deadline is the 14th July and not the 17th June as first stated. Please accept out apologies for any confusion caused.

Central moments

John Lee

ActEd Tutor
Staff member
Hi there

Just posting a question I received from a student as it's really good.

Hello, I am Ioannis , just a quick question on probability theory.

The variance is defined as E[(X - m)^2] and gives a measure of the spread of the returns around the mean.
Why is it wrong to take E [ l X - m l ] or E[(X - m)^4] instead as a measure of the spread?
All three expressions inside the brackets are +ve.

Similarly, the skewness is defined as E[(X - m)^3]. Why is it wrong to take E[(X - m)^5] instead?
What does E[(X - m)^5] represent and why?

I understand how the calculations work. I just want to see the logic behind the formulas.

Many thanks,

Ioannis
 
The variance is defined as E[(X - m)^2] and gives a measure of the spread of the returns around the mean.
Why is it wrong to take E [ l X - m l ] or E[(X - m)^4] instead as a measure of the spread?
All three expressions inside the brackets are +ve.

It's been a while since I've looked at these, but I would guess that E[|X-m|] is not used as it's messy to integrate - though I tried it out with f(x) = ¾(1-x²) -1<x<1 and got a value of 0.375 compared to a true sd of 0.447.

I think the variance also has more useful statistical properties such as the chi-square distribution whereas a modulus most definitely won't.

Got it! Just tried calculating E(|X-m|) for a U(0,2) and got zero! This will happen for all distributions that are wholly positive or wholly negative as E(|X-m|) = E(X-m) = E(X)-m = m - m = 0

The 4th central moment is the kurtosis of a distribution and doesn't give the spread but the curvature of the distribution. See here for details.

Similarly, the skewness is defined as E[(X - m)^3]. Why is it wrong to take E[(X - m)^5] instead?
What does E[(X - m)^5] represent and why?

I have no idea what the fifth moment gives - but looking at the progression it seems that it will give a different feature to the asymmetry. It still gives 0 for a symmetrical distribution but will be more weighted by extreme values.

General details about moments can be found here.
 
Last edited by a moderator:
There's a paragraph in the notes (Ch1 pg13) which refers to absolute function as "not a nice function to deal with mathematically - in particular ... has a nasty kink when x=0, whereas graph of x^2 is always a nice "smooth" curve."

A bit fluffy but gets the point across. It represents spread in the simplest way.

Higher order moments describe different things as bob says. E[(X - m)^5] represents the fifth central moment.
 
Back
Top