Since this returns a number between 0 and 1 (excluding 1 itself), my question is how can use this to represent a percentage. For example, if I choose a variable to be 0.75 and have it compare to the RNG, would there be a 25% of the result being higher? It seems to me it would be slightly less because it cannot return a value of 1. Or, maybe I would need to say there would be a 25% chance of it being equal to or greater than 0.75?
Using Random.NextDouble() method
- Thread starter Bvwalker1
- Start date
Skydiver
Staff member
My mind went elsewhere regarding the slight skewing of probabilities that happens as the result of using division to compute a random floating number. I'm glad @JohnH got to the heart of the matter.
Anyway, of interest to some people would be:
link.springer.com
For most practical purposes, it shouldn't matter, but if you are into deep numerical analysis and need truly evenly distributed values, then it would make a difference between comparing
Anyway, of interest to some people would be:
Generating Random Floating-Point Numbers by Dividing Integers: A Case Study
A method widely used to obtain IEEE 754 binary floating-point numbers with a standard uniform distribution involves drawing an integer uniformly at random and dividing it by another larger integer. We survey the various instances of the algorithm that are used...
link.springer.com
For most practical purposes, it shouldn't matter, but if you are into deep numerical analysis and need truly evenly distributed values, then it would make a difference between comparing
<[icode] vs. [icode]<=, as well as, switching to a different random number generator, or at least a different way of getting a random floating point number. (See reference source to see that C# uses the division method to a random floating point number.)Similar threads
- Question
- Question
