Statistical test, samples of very different sizes

  • Context: Undergrad 
  • Thread starter Thread starter Metis
  • Start date Start date
  • Tags Tags
    Statistical Test
Click For Summary
SUMMARY

The discussion centers on comparing two astronomical samples with significantly different sizes using statistical methods. The Kolmogorov–Smirnov test is identified as a suitable approach for determining if the two samples, which include parameters such as mass, period, eccentricity, and semi-major axis of planets orbiting different types of stars, come from the same probability distribution. The smaller sample size arises from observational challenges associated with one star type. The need for clarity on sample sizes and parameters is emphasized for accurate analysis.

PREREQUISITES
  • Understanding of the Kolmogorov–Smirnov test
  • Familiarity with statistical distributions
  • Knowledge of astronomical parameters such as mass and semi-major axis
  • Basic statistical analysis skills
NEXT STEPS
  • Research the Kolmogorov–Smirnov test in detail
  • Learn about other statistical tests for comparing samples of different sizes
  • Investigate methods for estimating parameters in small sample sizes
  • Explore the implications of sample size on statistical power and analysis
USEFUL FOR

Astronomers, statisticians, and researchers involved in comparative analysis of astronomical data, particularly those dealing with small sample sizes and distribution assessments.

Metis
Messages
6
Reaction score
0
So, I am working with a professor on some astronomy research. I need two compare two samples but one of them has a much smaller size due to the lack of observations.

The samples have about 5 parameters. I don't know much about statistics but from reading so far I think the Kolmogorov–Smirnov test would be the most appropriate.

Is that correct or is there any other tests I might want to look into? I need to see if the two samples are the same despite the fact that one is drawn from a smaller data set.

thanks!
 
Physics news on Phys.org
Do you mean you need to investigate the underlying distributions - do the two samples come from the same probability distribution?

Couple questions:
1) What are the sample sizes?
2) The parameters that need to be estimated - what are they?
 
1. The samples are planets orbiting different type stars. One of the types of stars has planets inherently harder to observe so its sample is smaller but I want to see if the distribution of the different types of planets for the two star types despite this difference
2. The parameters are things like mass, period, eccentrity, semi-major axis etc.
 
Last edited:

Similar threads

High School Chi-Square Tests for Homogeneity & Association
  • · Replies 7 ·
Replies
7
Views
3K
Undergrad Sampling theory and random sample
  • · Replies 5 ·
Replies
5
Views
3K
Undergrad Desirable estimator properties...
  • · Replies 7 ·
Replies
7
Views
3K
Undergrad Nonparametric Hypothesis Tests
  • · Replies 4 ·
Replies
4
Views
2K
High School AI Detection - Phase 1: sample collection
  • · Replies 4 ·
Replies
4
Views
2K
Undergrad Type 1 Error Increases with Sample Size?
  • · Replies 9 ·
Replies
9
Views
4K
Graduate 2 samples T test in case of non-normal distribution
  • · Replies 9 ·
Replies
9
Views
2K
Undergrad Two-sample t test vs. chi-squared test for homogeneity
  • · Replies 1 ·
Replies
1
Views
2K
Undergrad Proper understanding of p-value
  • · Replies 5 ·
Replies
5
Views
2K
Undergrad Linear regression and random variables
  • · Replies 30 ·
2
Replies
30
Views
5K