Create Gaussian Function from Coords: Learn Integration & Matrix

  • Thread starter Thread starter SHRock
  • Start date Start date
  • Tags Tags
    Function Gaussian
Click For Summary
SUMMARY

The discussion focuses on creating a Gaussian function from a set of coordinates using integration and matrix methods. The user has provided specific coordinates and a preliminary Gaussian function, y=0.01235140544*e^(-(x-36.28663863)^2/(-135.1643617)), and seeks to determine the parameters a, b, and c. A suggestion was made to apply least squares regression, particularly by taking the logarithm of the data to fit a second-degree polynomial, as the logarithm of a Gaussian is a polynomial function.

PREREQUISITES
  • Understanding of Gaussian functions and their properties
  • Familiarity with least squares regression techniques
  • Basic knowledge of integration methods
  • Experience with polynomial fitting and curve fitting techniques
NEXT STEPS
  • Learn about least squares regression and its application in curve fitting
  • Study the properties of Gaussian functions and their logarithmic forms
  • Explore integration techniques relevant to function approximation
  • Investigate matrix methods for solving systems of equations in curve fitting
USEFUL FOR

Students, mathematicians, and data analysts interested in statistical modeling, curve fitting, and those looking to apply mathematical functions to real-world data sets.

SHRock
Messages
8
Reaction score
0
I am completely new to this so bare with me.

I have co-ordinates and I want to create a Gaussian function. I already know the function which is

y=0.01235140544*e^(-(x-36.28663863)^2/(-135.1643617))

But I need to know how to get a,b,c

The Coordinates are

{35,.01}
{28,0.03}
{20,.1}
{15,.3}
{11,1}
{9,3}
{6,10}
{4.5,30}

Thanks

Edit: If your going to use integration then please give a basis or send a link to website where I can learn integration. Also is it possible to solve with a matrix. Well I don't really need a gaussian function, a power function will also work or any function with the closest curve fit
 
Last edited:
Physics news on Phys.org
SHRock said:
I am completely new to this so bare with me.

Well, I'm certainly not going to take off my clothes with you. :blushing:

I have co-ordinates and I want to create a Gaussian function. I already know the function which is

y=0.01235140544*e^(-(x-36.28663863)^2/(-135.1643617))

But I need to know how to get a,b,c

The Coordinates are

{35,.01}
{28,0.03}
{20,.1}
{15,.3}
{11,1}
{9,3}
{6,10}
{4.5,30}

Thanks

Edit: If your going to use integration then please give a basis or send a link to website where I can learn integration. Also is it possible to solve with a matrix. Well I don't really need a gaussian function, a power function will also work or any function with the closest curve fit

Hard to say. I suppose you want to do a least squares regression. Since the logarithm of a Gaussian is a second degree polynomial you might take the log of the data and try a least squares fit by a second degree polynomial.
 

Similar threads

  • · Replies 1 ·
Replies
1
Views
2K
Replies
2
Views
6K
  • · Replies 2 ·
Replies
2
Views
428