Reduced chi square for few data points

Click For Summary
SUMMARY

The discussion centers on fitting a straight line to a dataset consisting of 8 points with associated errors, specifically using reduced chi-squared analysis. The dataset includes x-values ranging from 1 to 8, y-values representing measurements, and y-errors reflecting Poisson errors. The user reports a reduced chi-squared value of 0.02, indicating a poor fit relative to the error margins. Participants emphasize the importance of understanding the origins of the error values and the potential impact of systematic errors on the analysis.

PREREQUISITES
  • Understanding of linear regression and curve fitting techniques
  • Familiarity with reduced chi-squared statistics
  • Knowledge of error propagation and Poisson statistics
  • Experience with data visualization tools for plotting datasets
NEXT STEPS
  • Investigate methods for improving linear fit accuracy in datasets with large error margins
  • Learn about systematic errors and their effects on statistical analysis
  • Explore advanced fitting techniques using tools like SciPy's curve_fit
  • Study the implications of reduced chi-squared values in model evaluation
USEFUL FOR

Researchers, data analysts, and statisticians involved in data fitting and error analysis, particularly those working with small datasets and seeking to understand the implications of statistical measures like reduced chi-squared.

Malamala
Messages
348
Reaction score
28
Hello! I need to make a straight line fit to 8 points, with errors on them. The data is like this ##x = [1,2,3,4,5,6,7,8]##, ##y=[377.488 691.191 , 1030.319, 1428.801, 1753.884, 2113.065 , 2398.642, 2797.664]##, ##y_{err}=[97.145, 131.452, 160.492, 188.997, 209.397, 229.840, 244.879, 264.464]##. The problem is that the errorbars (which are basically Poisson errors) are a lot bigger than the spread of the data. Hence I end up with a reduced chi-squared of 0.02. Is there anything I can do? The values of the fits are reasonable (given theoretical arguments), but I am not sure what to make out of this small reduced chi-square.
 
Physics news on Phys.org
Did you make a plot ?
Tell us how the data were obtained and what they represent. Especially the ##y_{err}##.
And how can you obtain such unbelievably accurate estimates of ##y_{err}##. Billions of observations, or just mindless copying calculation results ?
Are you aware of the role of systematic errors ?

BvU said:
How are the other threads going ? Long forgotten ?
I'd like to know if you understood the help given in earlier threads, or just lost interest and never reacted any more. I don't mind repeating things, but it should have a reasonable purpose.
 
Last edited:

Similar threads

  • · Replies 11 ·
Replies
11
Views
3K
  • · Replies 5 ·
Replies
5
Views
5K
  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 8 ·
Replies
8
Views
2K
  • · Replies 3 ·
Replies
3
Views
4K
  • · Replies 5 ·
Replies
5
Views
9K
  • · Replies 1 ·
Replies
1
Views
4K
  • · Replies 26 ·
Replies
26
Views
3K
  • · Replies 8 ·
Replies
8
Views
7K
  • · Replies 5 ·
Replies
5
Views
4K