Calculating Error in Linear Function b

In summary, the conversation includes a discussion about calculating the error in b when using a statistical method to interpolate a linear function. The person is using a program called "curve expert 1.3" to do the interpolation and they mention a law and an image that is attached. The other person suggests using a least square fit method, but the first person explains that their program is not using that method. They also mention not being able to provide detailed help due to the size of the file.
  • #1
TheDestroyer
402
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
  • #2
Are you using "least square fit" ?
 
  • #3
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: 372
Last edited:
  • #4
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.
 

What is the formula for calculating error in linear function b?

The formula for calculating error in linear function b is:
b_error = (Σ(y-mean(y))(x-mean(x))) / (Σ(x-mean(x))^2)

What do each of the variables in the error calculation formula represent?

The variables in the error calculation formula represent the following:
b_error: the error in the slope of the linear function
Σ: the sum of
y: the y-values of the data points
mean(y): the mean of the y-values
x: the x-values of the data points
mean(x): the mean of the x-values

What does a positive or negative value for b_error indicate?

A positive value for b_error indicates that the slope of the linear function is overestimated, while a negative value indicates an underestimation. In other words, a positive b_error means that the line is steeper than it should be, while a negative b_error means that the line is less steep than it should be.

How do you interpret the magnitude of b_error?

The magnitude of b_error represents the amount of error in the slope of the linear function. A larger magnitude indicates a larger error, while a smaller magnitude indicates a smaller error. It is important to consider the magnitude of b_error in relation to the range of the data and the overall accuracy of the linear function.

Can b_error be used to determine the accuracy of the entire linear function?

No, b_error only represents the error in the slope of the linear function. To determine the overall accuracy of the linear function, other measures such as the coefficient of determination (R-squared) and residual plots should also be considered.

Similar threads

  • General Math
Replies
5
Views
1K
Replies
8
Views
2K
Replies
3
Views
194
  • General Math
Replies
31
Views
1K
  • General Math
Replies
2
Views
714
Replies
17
Views
2K
  • General Math
Replies
2
Views
9K
Replies
2
Views
1K
Replies
4
Views
1K
Replies
10
Views
1K
Back
Top