Difference between revisions of "Mdm of a discrete distribution lemma"

From Maths
Jump to: navigation, search
m (link typo)
m (Let's try a new version)
 
(One intermediate revision by the same user not shown)
Line 1: Line 1:
 
I may have found a useful transformation for calculating [[Mdm|Mdm's]] of distributions defined on {{M|\mathbb{Z} }} or a subset. I document my work so far below:
 
I may have found a useful transformation for calculating [[Mdm|Mdm's]] of distributions defined on {{M|\mathbb{Z} }} or a subset. I document my work so far below:
 
:* [[Notes:Mdm of a discrete distribution lemma]]
 
:* [[Notes:Mdm of a discrete distribution lemma]]
 +
:* [[Notes:Mdm of a discrete distribution lemma - round 2]]{{ProbMacros}}
 +
__TOC__
 +
==Statement==
 +
:: {{Notice|'''Notice: ''' - there are plans to generalise this lemma:- specifically to allow {{M|\lambda}} to take any real value (currently only non-negative allowed) and possibly also allow {{M|\alpha,\beta}} to be negative too}}
 +
Let {{M|\lambda\in\mathbb{R}_{\ge 0} }} and let {{M|\alpha,\beta\in\mathbb{N}_0}} such that {{M|\alpha\le\beta}}, let {{M|f:\{\alpha,\alpha+1,\ldots,\beta-1,\beta\}\subseteq\mathbb{N}_0\rightarrow\mathbb{R} }} be a [[function]], then we claim:
 +
* {{MM|\sum^\beta_{k\eq\alpha}\big\vert k-\lambda\big\vert f(k) \eq\sum^\gamma_{k\eq\alpha}(\lambda-k)f(k)  +\sum_{k\eq\gamma+1}^\beta (k-\lambda)f(k) }} where:
 +
** {{M|\gamma:\eq\text{Min}(\text{Floor}(\lambda),\beta)}}
 +
Note that {{M|\beta\eq\infty}} is valid for this expression (standard limits stuff, see ''[[sum to infinity]]'')
 +
==Applications to computing [[Mdm]]==
 +
Let {{M|X}} be a [[discrete random variable|''discrete'']] [[random variable]] defined on {{M|\{\alpha,\alpha+1,\ldots,\beta-1,\beta\}\subseteq\mathbb{N}_0}} (remember that {{M|\beta\eq\infty}} is valid and just turns the second sum into a limit), then:
 +
* define {{M|\lambda:\eq\E{X} }}
 +
* define {{M|f:k\mapsto \P{X\eq k} }}
 +
 +
Recall the [[mdm]] of {{M|x}} is defined to be:
 +
* {{MM|\Mdm{X}:\eq \sum^\beta_{k\eq\alpha}\big\vert k-\E{X}\big\vert\ \P{X\eq k} }}
 +
It is easy to see that with the definitions substituted that:
 +
* {{MM|\sum^\beta_{k\eq\alpha}\big\vert k-\lambda\big\vert f(k)\eq\Mdm{X} }}
 +
==Proof==
 +
{{Requires proof|grade=A**|msg=Note follow
 +
* Initial notes [[User:Alec|Alec]] ([[User talk:Alec|talk]]) 01:24, 22 January 2018 (UTC)
 +
*# A lot of work has been done in ''[[Notes:Mdm of a discrete distribution lemma]]'' and I've done each of the 4 cases individually ({{M|\alpha\eq\beta}}, {{M|\beta<\text{Floor}(\lambda)}}, {{M|\beta>\text{Floor}(\lambda)}} and {{M|\beta\eq\text{Floor}(\lambda)}} - but they need to be put together.
 +
*# There is a 5th case where {{M|\lambda<0}} is introduced
 +
*# I'd like to generalise this to {{M|\alpha,\beta\in\mathbb{Z} }} - generalising beyond {{M|\alpha,\beta}} being ''non-negative''}}
 +
==Notes==
 +
<references group="Note"/>
 +
==References==
 +
<references/>
 
{{Theorem Of|Statistics|Probability|Elementary Probability}}
 
{{Theorem Of|Statistics|Probability|Elementary Probability}}

Latest revision as of 17:53, 24 January 2018

I may have found a useful transformation for calculating Mdm's of distributions defined on Z or a subset. I document my work so far below:

  • Notes:Mdm of a discrete distribution lemma
  • Notes:Mdm of a discrete distribution lemma - round 2

Statement

Notice: - there are plans to generalise this lemma:- specifically to allow λ to take any real value (currently only non-negative allowed) and possibly also allow α,β to be negative too

Let λ∈R≥0 and let α,β∈N0 such that α≤β, let f:{α,α+1,…,β−1,β}⊆N0→R be a function, then we claim:

  • β∑k=α|k−λ|f(k)=γ∑k=α(λ−k)f(k)+β∑k=γ+1(k−λ)f(k) where:
    • γ:=Min(Floor(λ),β)

Note that β=∞ is valid for this expression (standard limits stuff, see sum to infinity)

Applications to computing Mdm

Let X be a discrete random variable defined on {α,α+1,…,β−1,β}⊆N0 (remember that β=∞ is valid and just turns the second sum into a limit), then:

  • define λ:=E[X]
  • define f:k↦P[X=k]

Recall the mdm of x is defined to be:

  • Mdm(X):=β∑k=α|k−E[X]| P[X=k]

It is easy to see that with the definitions substituted that:

  • β∑k=α|k−λ|f(k)=Mdm(X)

Proof

Grade: A**
This page requires one or more proofs to be filled in, it is on a to-do list for being expanded with them.
Please note that this does not mean the content is unreliable. Unless there are any caveats mentioned below the statement comes from a reliable source. As always, Warnings and limitations will be clearly shown and possibly highlighted if very important (see template:Caution et al).
The message provided is:
Note follow
  • Initial notes Alec (talk) 01:24, 22 January 2018 (UTC)
    1. A lot of work has been done in Notes:Mdm of a discrete distribution lemma and I've done each of the 4 cases individually (α=β, β<Floor(λ), β>Floor(λ) and β=Floor(λ) - but they need to be put together.
    2. There is a 5th case where λ<0 is introduced
    3. I'd like to generalise this to α,β∈Z - generalising beyond α,β being non-negative

Notes

References


Retrieved from "http://www.maths.kisogo.com/index.php?title=Mdm_of_a_discrete_distribution_lemma&oldid=9230"