Continuous-time processes with attraction?

  • Context: Graduate 
  • Thread starter Thread starter CRGreathouse
  • Start date Start date
  • Tags Tags
    Attraction
Click For Summary
SUMMARY

The discussion centers on finding a suitable statistical model for continuous-time processes that exhibit attraction or repulsion among points, specifically in the context of data that does not conform to the independence assumption of the Poisson process. The negative binomial distribution is proposed as a potential alternative, as it includes a dispersion parameter that may allow for adjustments in the degree of clumping or attraction among data points. Participants express interest in testing this model to evaluate its effectiveness in their specific applications.

PREREQUISITES
  • Understanding of Poisson processes and their independence assumption
  • Familiarity with negative binomial distribution and its parameters
  • Basic knowledge of statistical modeling techniques
  • Experience with data analysis and interpretation
NEXT STEPS
  • Research the properties and applications of the negative binomial distribution
  • Explore statistical modeling techniques for continuous-time processes
  • Learn about the implications of the independence assumption in statistical models
  • Investigate alternative models that incorporate attraction or repulsion, such as point process models
USEFUL FOR

Statisticians, data analysts, and researchers seeking to model complex interactions in data sets, particularly those dealing with phenomena that exhibit clustering or spatial dependencies.

CRGreathouse
Science Advisor
Homework Helper
Messages
2,832
Reaction score
0
I'm trying to find a good model, and I wanted to find a process that fits. The Poisson process almost fits, but unfortunately the independence assumption is too strong for my data. Is there a similar process that has a parameter (or several) that allow for points to attract or repel others, like zeta zeros?

I've never been good with statistics, so there could be a very simple process I've overlooked, I don't know.
 
Physics news on Phys.org
CRGreathouse said:
I'm trying to find a good model, and I wanted to find a process that fits. The Poisson process almost fits, but unfortunately the independence assumption is too strong for my data. Is there a similar process that has a parameter (or several) that allow for points to attract or repel others, like zeta zeros?

I've never been good with statistics, so there could be a very simple process I've overlooked, I don't know.

Negative binomial distribution?
The distribution is described by 2 parameters.
One could be considered a "dispersion" parameter;
a degree of "clumping" in the population?
Maybe it could provide a knob for dialing up or down the degree of attraction or repulsion.
 
fopc said:
Negative binomial distribution?
The distribution is described by 2 parameters.
One could be considered a "dispersion" parameter;
a degree of "clumping" in the population?
Maybe it could provide a knob for dialing up or down the degree of attraction or repulsion.

Oh, cool. This looks good -- I'll have to try it with my population to see how well it can model it.

Thanks!
 

Similar threads

  • · Replies 12 ·
Replies
12
Views
4K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 4 ·
Replies
4
Views
3K
  • · Replies 7 ·
Replies
7
Views
3K
  • · Replies 9 ·
Replies
9
Views
3K
  • · Replies 13 ·
Replies
13
Views
2K
  • · Replies 10 ·
Replies
10
Views
3K
  • · Replies 1 ·
Replies
1
Views
3K
  • · Replies 1 ·
Replies
1
Views
716
  • · Replies 1 ·
Replies
1
Views
3K