Calculating Error in Linear Function b

  • Context: Undergrad 
  • Thread starter Thread starter TheDestroyer
  • Start date Start date
  • Tags Tags
    Errors Function Linear
Click For Summary

Discussion Overview

The discussion revolves around calculating the error in the parameter 'b' of a linear function derived from experimental data. Participants explore the methods used for interpolation and error calculation, focusing on the statistical approach rather than least squares fitting.

Discussion Character

  • Exploratory, Technical explanation, Debate/contested

Main Points Raised

  • One participant describes their linear function as y=(a/b^2)x+c and seeks to understand how to calculate the error in 'b' given the standard error in 'y' (S).
  • Another participant inquires whether the original poster is using least squares fitting for their analysis.
  • The original poster clarifies that they are using a statistical method and provides details about their approach, including the use of a specific program for interpolation.
  • A later reply suggests that many programs, including the one mentioned, often utilize least squares fitting, but expresses inability to assist further due to unfamiliarity with the program used by the original poster.

Areas of Agreement / Disagreement

Participants do not reach a consensus on the method of error calculation, as there is a divergence in approaches between statistical methods and least squares fitting.

Contextual Notes

The discussion includes assumptions about the statistical method being used and the nature of the program for interpolation, which may not be universally applicable or understood by all participants.

TheDestroyer
Messages
401
Reaction score
1
Hi,

When I'm doing an experiment,

If the result of the experiment was some points, which I interpolate to create the linear function:

y=(a/b^2) x+c

while y is a function of x: y=f(x)

If the standard error in y was (S), How can I calculate the error in b?

Thank you
 
Mathematics news on Phys.org
Are you using "least square fit" ?
 
No, I'm using the statistical way, you can see its law in the image I've attached,

(((n_points is the quantity of the used points, y_i is the y-coordinate of the point i due to experiment, f(x_i) is the y-coordinate of the point i due to the interpolated function, n_param is the repeat of every point which equals always one as I think in my experiment because every point is measured once)))

I'm using program named "curve expert 1.3" to interpolate the linear function, you can download it for free and check the errors details, i was not able to upload the detailed help file because it's big

Is this information enough?
 

Attachments

  • StandardError.jpg
    StandardError.jpg
    18.3 KB · Views: 446
Last edited:
Well these programs are very often doing least square fit.. but the calculate it very fast hehe.

Least square fit is a statistical method.

Scince you are using a program that i don't use, i can't help you. sorry.
 

Similar threads

  • · Replies 6 ·
Replies
6
Views
2K
  • · Replies 6 ·
Replies
6
Views
3K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 1 ·
Replies
1
Views
1K
  • · Replies 2 ·
Replies
2
Views
16K
  • · Replies 19 ·
Replies
19
Views
2K
Replies
3
Views
2K
  • · Replies 1 ·
Replies
1
Views
1K
  • · Replies 21 ·
Replies
21
Views
3K
  • · Replies 14 ·
Replies
14
Views
3K