Lecture_Statistics_Spring_2013b

# Propagation of error example 1 consider the perimeter

This preview shows page 1. Sign up to view the full content.

This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: ȹ ȹ ȹ ∂X ȹ ȹ ȹ ∂X1 Ⱥ X 2 ,…, X N ȹ ∂X 2 Ⱥ X1 , X 3 ,…, X N ȹ N Ⱥ X1 , X N −1 2) Square of above Equation: Ⱥ Ⱥ ȹ ∂Q ȹ ȹ ∂Q ȹ 2 Ⱥȹ ∂Q ȹ (dQ ) = ȹ ȹ dX1 + ȹ dX 2 + … + ȹ dX N Ⱥ ȹ ∂X ȹ ȹ ∂X ȹ ȹ ȹ ∂X ȹ ȹ Ⱥȹ 1 Ⱥ X ,…, X Ⱥ ȹ 2 Ⱥ X1 , X 3 ,…, X N ȹ N Ⱥ X1 , X N−1 2 N Ⱥ Ⱥ 2 IV. Propagation of Error 2 σ Q = (dQ )2 = 2 Ⱥȹ ∂Q ȹ 2 ȹ ȹ ∂Q ȹ Ⱥȹ ȹ ȹ (dX 2 )2 + … + ȹ ∂Q (dX1 ) 2 + ȹ ȹ ∂X ȹ ȹ ∂X Ⱥȹ ∂X1 ȹ X ,…, X ȹ Ⱥ 2 ȹ 2 Ⱥ X1 , X 3 ,…, X N ȹ N N Ⱥ 2 Ⱥ ȹ 2 Ⱥ ȹ (dX N ) ȹ Ⱥ Ⱥ X1 , X N −1 Ⱥ Noting that 1) the dXi2 are variances, i.e. s2Xi, 2) the dX1, dX2, etc. are sX1, sX2, etc., respectively, 3) dX1dX2, dX1dX3, etc. are covariances, which are zero if the Xi are independent. 2 2 Ⱥȹ ∂Q ȹ 2 Ⱥ ȹ ∂Q ȹ ȹ ∂Q ȹ 2 2 2 2 Ⱥ σ Q = Ⱥȹ σ X + ȹ σ X + … + ȹ σX ȹ ∂X ȹ ȹ ȹ ∂X ȹ ȹ ȹ ∂X ȹ ȹ 1 2 N Ⱥ Ⱥȹ 1 Ⱥ X ,…, X ȹ 2 Ⱥ X1 , X 3 ,…, X N ȹ N Ⱥ X1 , X N−1 2 N Ⱥ Ⱥ Eq* σQ = N ȹ ∂Q ȹ ∑ ȹ ȹ i =1ȹ X i 2 ȹ σ 2 ȹ Xi Ⱥ X j *j represents all other N – 1 independent variables IV. Propagation of Error Example 1 Consider the perimeter of a right triangle which has...
View Full Document

## This note was uploaded on 01/26/2014 for the course CHEM 3625 taught by Professor Mrjohnson during the Spring '08 term at Virginia Tech.

Ask a homework question - tutors are online