Lecture_16_regression_F07

Lecture_16_regression_F07 - 1 E7: INTRODUCTION TO COMPUTER...

Info iconThis preview shows pages 1–10. Sign up to view the full content.

View Full Document Right Arrow Icon
E7 L15 1 E7: INTRODUCTION TO COMPUTER E7: INTRODUCTION TO COMPUTER PROGRAMMING FOR SCIENTISTS AND PROGRAMMING FOR SCIENTISTS AND ENGINEERS ENGINEERS Lecture Outline 1. Least squares solution when n = # equations >> m = # of unknowns 1. Regression and curve fit Copyright 2007, Horowitz, Packard. This work is licensed under the Creative Commons Attribution-Share Alike License. To view a copy of this license, visit http://creativecommons.org/licenses/by-sa/2.0/ or send a letter to Creative Commons, 559 Nathan Abbott Way, Stanford, California 94305, USA.
Background image of page 1

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

View Full DocumentRight Arrow Icon
E7 L15 2 Linear equations in matrix form Linear equations in matrix form Consider n LINEAR equations and m unknowns. A   matrix ( n x m ) p   vector ( m x 1) y   vector ( n x 1) unknown
Background image of page 2
E7 L15 3 Linear equations in matrix form Linear equations in matrix form = n-by- m m -by-1 n-by-1 = n-by- m m -by-1 n-by-1 A p y A p y
Background image of page 3

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

View Full DocumentRight Arrow Icon
E7 L15 4 Linear equations in matrix form when Linear equations in matrix form when n n >> >> m m = n-by- m m -by-1 n-by-1 A p y n n : number of equations is much larger than m m : number of unknowns
Background image of page 4
E7 L15 5 The backslash operator: The backslash operator: p = A\y Given a set of n linear equations with m unknowns   Computes p that solves : >> p = A\y solves the least squares problem.
Background image of page 5

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

View Full DocumentRight Arrow Icon
E7 L15 6 The backslash operator: The backslash operator: p = A\y What happens if the least squares (LS) solution is not unique and n n  ≠   ≠  m m ? >> p = A\y computes a basic LS solution, which has at most r nonzero components (When n n >> >> m m , this is normally not the case…) # of columns of A # of LI columns of A when r r  <   <  m m
Background image of page 6
E7 L15 7 -8 -6 -4 -2 0 2 4 6 8 10 -6 -5 -4 -3 -2 -1 0 2 Find the linear function: that minimizes the sum of error squares: Linear Regression: Curve-fitting with minimum error Linear Regression: Curve-fitting with minimum error Given n  (x,y) data pairs: (x 1 ,y 1 ), (x 2 ,y 2 ), … ,  (x n ,y n ) y x
Background image of page 7

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

View Full DocumentRight Arrow Icon
E7 L15 8 Linear Regression: Computing the LS solution Linear Regression: Computing the LS solution Given n  (x,y) data pairs: (x 1 ,y 1 ), (x 2 ,y 2 ), … ,  (x n ,y n ), For a given set of parameters: p 1 and p 2   , we can compute the errors e i ’s  as follows
Background image of page 8
E7 L15 9 Linear Regression: Computing the LS solution Linear Regression: Computing the LS solution Given n  (x,y) data pairs: (x 1 ,y 1 ), (x 2 ,y 2 ), … ,  (x n ,y n ), In vector/matrix form: p   vector ( 2 x 1) A   matrix ( n x 2 ) y   vector ( n x 1) e   vector ( n x 1)
Background image of page 9

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

View Full DocumentRight Arrow Icon
Image of page 10
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 30

Lecture_16_regression_F07 - 1 E7: INTRODUCTION TO COMPUTER...

This preview shows document pages 1 - 10. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online