Method for experimental results analysis

Click For Summary

Discussion Overview

The discussion revolves around methods for analyzing experimental results involving multiple variables and observations. Participants explore different approaches to create a unified function from the results obtained in the experiment.

Discussion Character

  • Exploratory
  • Technical explanation
  • Mathematical reasoning

Main Points Raised

  • One participant describes their experimental setup with three variables and two observation results, seeking better analysis methods.
  • Another participant suggests that the analysis depends on the expected mathematical relationship and recommends using least squares fitting once possible functions are nominated.
  • A different participant proposes a linear relationship model, suggesting the form y₁ = a·x₁ + b·x₂ + c·x₃, and mentions solving for parameters through numerical methods.
  • Another participant offers a visualization approach, recommending a 3D graph to represent the variables and observations, with suggestions for using color and texture to convey information about the dependent variables.

Areas of Agreement / Disagreement

Participants express various methods and ideas for analysis and visualization, but there is no consensus on a single approach or method. Different perspectives on the mathematical relationships and visualization techniques remain present.

Contextual Notes

Participants have not fully defined the assumptions regarding the mathematical relationships or the specific nature of the data, which may affect the proposed methods.

miraboreasu
Messages
24
Reaction score
0
Hello guys, I have conducted an experiment and got some results.
I have 3 variables to vary, for example, five x1, five x2, and two x3
and 2 observation results, like y1, y2
I already make y1 y2 and x1 x2 x3 dimensionless
since plot is 2D, what I am doing now is just plot when x3=1, x2=1, plot the first y1(x1), x3=1, x2=2, plot the second y1(x1), etc
the same plots for y2

I want to know if is there a better way to analyze the results, for example, what method I should learn to create a function y1=(x1,x2,x3)?
 
Physics news on Phys.org
Analysis will depend on what mathematical relationship you are expecting. Once you nominate the possible functions and parameters, you can solve for those parameters using a least squares fit.
 
  • Like
Likes   Reactions: topsquark
Baluncore said:
Analysis will depend on what mathematical relationship you are expecting. Once you nominate the possible functions and parameters, you can solve for those parameters using a least squares fit.
Thanks, sir, but I didn't quite understand, I have multi varying, and I want to create a unified function for them.

What I am thinking is y1= alpha x1+beta x1+gamma x1. But I want to know more details
 
Last edited:
I'm not real sure what your desired end result is, but if you are trying to get 3 variables and 1 or 2 results on a single graph, here is one approach:

1) A surface represented as a 3D graph with a variable on each axis
2) the dependent variable(s) (y1, y2) as the color of the surface
3) the color could be a mix to show the various mixes of y1, y2; or maybe color and texture... hmm... even shading lines, say density or line width in one axis for y1 and in a cross direction for y2.

A 3D bar graph is another possible representation which may be better if you need to print the result, rather than interactively display it on-screen.

It has been decades, but I seem to recall that programs like Microsoft Office can show 3D graphs. Of course there are many free Office-type programs now available.

Please let us know what you come up with. We like to learn too!

Cheers,
Tom
 
  • Like
Likes   Reactions: miraboreasu and topsquark

Similar threads

  • · Replies 4 ·
Replies
4
Views
1K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 5 ·
Replies
5
Views
3K
  • · Replies 14 ·
Replies
14
Views
4K
  • · Replies 1 ·
Replies
1
Views
985
  • · Replies 13 ·
Replies
13
Views
2K
  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 9 ·
Replies
9
Views
2K
  • · Replies 6 ·
Replies
6
Views
5K
  • · Replies 2 ·
Replies
2
Views
2K