Error on slope using LSQ method

Click For Summary

Homework Help Overview

The problem involves calculating the error on the slope using the least squares (LSQ) method based on three given points with associated errors in their y-coordinates. The participants are discussing the relevant equations and calculations necessary for this analysis.

Discussion Character

  • Mathematical reasoning, Assumption checking

Approaches and Questions Raised

  • The original poster attempts to apply the LSQ method but expresses uncertainty about the correctness of the equations used and the calculations performed. Some participants question the formulation of the equation for σy, particularly regarding the placement of the minus sign. Others seek clarification on the meaning of y(xk) in the context of the problem.

Discussion Status

The discussion is ongoing, with participants actively engaging in clarifying the equations and addressing potential errors in the original poster's calculations. There is a focus on ensuring the correct interpretation of the formulas and the variables involved.

Contextual Notes

There are indications of confusion regarding the mathematical expressions and their implications, particularly concerning the calculation of Δ and the interpretation of y(xk). The original poster's calculations have led to negative values, raising questions about the validity of the approach taken.

Matt21
Messages
14
Reaction score
0

Homework Statement


Three points are given with errors on y coordinates.
(2.00,4.70±0.36); (4.00,6.8&±0.45); (5.00, 9.10±0.50)
Using LSQ method, find the error on slope.

Homework Equations


σm = σy√(n/Δ), where σy = √(1/n-2(Σ(yk - y(xk))^2)) and Δ = nΣxk^2 - (Σxk)^2

The Attempt at a Solution


I do not know if this is the correct equation or if I made an error in my calculations but what I got was:
σy = 1/3-2((4.7-4.7(2))^2+(6.8-6.8(4))^2+(9.1-9.1(5))^2) = 42 and
Δ = (2^2+4^2+5^2)-(2+4+5)^2 = -76
n = 3
Obviously I can't use Δ since you can't get a root of a negative number. Can anyone see what I'm doing wrong? Any help would be much appreciated.
 
Physics news on Phys.org
Matt21 said:
1/n-2(Σ(yk - y(xk))^2))
That's a very strange looking expression. Are you sure about the minus sign after 1/n?
 
Yes I'm positive. The formula given in my notes was √((1/n-2)(Σ(yk - y(xk))^2)
 
Matt21 said:
Yes I'm positive. The formula given in my notes was √((1/n-2)(Σ(yk - y(xk))^2)
Ah, you mean √(1/(n-2)(Σ(yk - y(xk))^2)), i.e. ##\sqrt{\frac 1{n-2}\Sigma( y_k-y(x_k))^2}##? That would make more sense. Originally you had √((1/n)-2Σ(yk - y(xk))^2)), i.e. ##\sqrt{\frac 1n-2\Sigma(y_k-y(x_k))^2}##.
By the way, please clarify what you mean by y(xk).
 
Yes that was the equation was trying to convey. In regards to y(xk) I would assume that means y*(xk) where xk = x1, x2, x3...xk
 
Last edited:
Δ = (3*(2^2+4^2+5^2))-(2+4+5)^2 = 14
 

Similar threads

  • · Replies 8 ·
Replies
8
Views
4K
  • · Replies 13 ·
Replies
13
Views
3K
  • · Replies 3 ·
Replies
3
Views
2K
Replies
9
Views
3K
  • · Replies 65 ·
3
Replies
65
Views
9K
  • · Replies 9 ·
Replies
9
Views
2K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 9 ·
Replies
9
Views
3K
  • · Replies 1 ·
Replies
1
Views
4K
  • · Replies 15 ·
Replies
15
Views
3K