Random Numerical Analysis

Many programs require the generation of random numbers.

Many computer programs need to generate random numbers properly. How random those generated numbers are determines how effective the programs are. Random number analysis looks at methods to determine how random the numbers they generate are.

Standard Methods

Programming languages often come with built-in functions for generating random numbers. However, these are simple numbers and often deterministic. Namely, they take an input, such as the computer's current time, and feed it into an equation which produces a number.

More Advanced Methods

More advanced methods of generating random numbers include translating more random inputs, such as processor noise, into quantitative inputs. These inputs are then plugged into equations and used to produce more random numbers.

Analysis

Random number analysis examines the numbers produced by a given method to determine how random they are. This is often done by looking for patterns in the numbers the methods generate. The less of a pattern the method follows, the more random it is. One way of doing this is to produce outputs into visual representations of a method's outputs, and looking at it for patterns.

References

• The Python Standard Library: Random - Generate Pseudo-Random Numbers
• Random.org: Statistical Analysis