Error Propagation - Estimating Variance

mardi 29 octobre 2013

1. The problem statement, all variables and given/known data



Not exactly a homework question, but rather a section in Statistical Data Analysis:



Suppose there is a pdf y[SUB](x)[/SUB] that is not completely known, but μi and Vij are known:



2. Relevant equations







3. The attempt at a solution



I understand how <y(x) > ≈ y(μ),



My confusion:



Why does <y(x2) >



1. Imply we square everything throughout?



<[y(μ) + Ʃ[∂y/∂x](xi - μi)]2>



2. give a xi and xj term? Where did the xj come from?



3. Why is it for i≠j when xi and xj are uncorrelated, the expression simplifies to



σ2y ≈ Ʃ[∂y/∂x]2σ2i



Where did the j go?











via Physics Forums RSS Feed http://www.physicsforums.com/showthread.php?t=719631&goto=newpost

0 commentaires:

Enregistrer un commentaire