Random number with which distribution

Click For Summary
SUMMARY

The discussion focuses on generating random numbers from a distribution where each preceding item has a higher probability than the subsequent ones, specifically for the range of 1 to 8. The user currently employs a normal distribution with a mean of 1 and varying sigmas but finds this method academically unacceptable. They seek a more appropriate distribution that allows for slope adjustment and proper parameterization. The user has also attempted using a Poisson distribution but encountered challenges in controlling the probability drop-off and constraining the output range.

PREREQUISITES
  • Understanding of probability distributions, specifically normal and Poisson distributions.
  • Familiarity with parameters such as mean, sigma, and lambda in statistical contexts.
  • Knowledge of random number generation techniques in programming.
  • Basic concepts of discrete probability and its applications.
NEXT STEPS
  • Research the use of the exponential distribution for controlling probability slopes.
  • Explore the beta distribution for flexible parameterization and bounded outputs.
  • Investigate the use of weighted random sampling techniques to achieve desired probabilities.
  • Learn about the implementation of custom probability distributions in programming languages like Python or R.
USEFUL FOR

Statisticians, data scientists, and researchers looking to generate controlled random variables with specific probability distributions for simulations or academic purposes.

wmac
Messages
19
Reaction score
0
Hello,

I am trying to find a clean solution for this (i.e. straightforward and academically acceptable). I appreciate if someone can help me.

1- I need to select between selections 1 to n (in my case 1-8 ). However each preceding item should have higher probability than the next one. For example, first item should have higher probability than the 2nd and the 2nd should have higher than the 3rd one etc.

I also need a parameter which enables me to adjust the probability of the first item. (i.e. the slope of the curve and therefore the distribution of probability)

Capture.JPG


What distribution with what parameters I should use.

2- Currently I use a normal distribution with mean value of 1 and different sigmas. When generating random numbers (float) I ignore numbers lower than 1 and then round up numbers higher than that to my discrete choices (i.e. 1-8 ).

Describing this in an academic paper is unacceptable I guess. So I am looking for a proper clean method (ideally a specific distribution with a few parameters).

Thanks in advance.
 
Physics news on Phys.org
I have not received any answer for this question.

I need to know whether this means there is no specific distribution for my purpose?

-----
BTW I have previously tried to do this with Poisson distribution but there are 2 problems:

- How can I control the slope (considering that I should use lambda=1). I mean I want to be able to control, how quickly the probability drops from one item to the next.

- In addition, how can I force the range of 0-7 (or 1-8 ) to the generated numbers?

Thanks
 

Similar threads

Replies
15
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 11 ·
Replies
11
Views
3K
  • · Replies 11 ·
Replies
11
Views
3K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 1 ·
Replies
1
Views
1K
  • · Replies 6 ·
Replies
6
Views
3K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 7 ·
Replies
7
Views
2K