Quiz3Soln

Quiz3Soln - ECM 005 Quiz 3 Solutions D.M Lavenson March 1...

This preview shows pages 1–3. Sign up to view the full content.

ECM 005 Quiz 3 Solutions D.M. Lavenson March 1, 2011 1 Problem 1 This problem can be approached the same we attemped the error analysis problem in Problem Set 11, though we have a slightly different functionality for y . We know that we have a data set which spans up to N points, so we will find an expression for the value of A that minimizes the square error between the data and the function y . We start first by defining the error and the square error between the data points y i and the predicted values A * f ( t i ) . E i = N s i =1 ( y i - Aln ( t i )) (1) E 2 i = N s i =1 ( y i - Aln ( t i )) 2 = N s i =1 y 2 i - 2 * A N s i =1 ( y i ln ( t i )) + A 2 N s i =1 ln ( t i ) 2 (2) Recall that each separate term must have the same summation notation applied to it. In order to minimize the squared error, we take the derivative of E 2 i with respect to A and set it equal to 0. dE 2 i dA = - 2 N s i =1 ( y i ln ( t i )) + 2 * A N s i =1 ln ( t i ) 2 = 0 (3) This equation is not our final answer! We must solve for an expression for A . Thus we rearrange to solve for A and we get a final expression defining A when minimizng the 1

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
square error of the data and predicted function y . A
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 11/11/2011 for the course ECM 5 taught by Professor Gates,b during the Spring '08 term at UC Davis.

Page1 / 5

Quiz3Soln - ECM 005 Quiz 3 Solutions D.M Lavenson March 1...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online