Random-Number-Generator

RNG generating numbers from various distributions along with:

• Series tests that check if numbers are generated randomly by constructing another series and comparing numbers of series in input with median value
• Chi-Square tests that check if generated numbers are of right distribution by comparing observed values with estimated values.

For specific and detailed information about the project please read the documentation [POLISH].

---

There are seven generators implemented:

Generator	Description
G	generates random integers
J	generates random fractions in range (0;1)
B	generates numbers with Bernoulli distribution
D	generates numbers with binomial distribution
P	generates numbers with Poisson distribution
W	generates numbers with exponential distribution
N	generates numbers with normal distribution

Generator J is constructed using G. B, D, P, W and N are all constructed using J.

---

Tests outcome:

I have tested every generator with series tests and the results were as expected: every generator except B and P generates correct pseudo-random numbers. Bernoulli cannot generate "very" random numbers due to the fact that is only generates '0' of '1'. Poisson on the other hand, has a lot duplicate values. I have performed Chi-Square tests only for generators B and P. Results for every data set were satisfying.

Test for 1000 generated numbers	G	J	B	D	P	W	N
Series test	✔️	✔️	❌	✔️	❌	✔️	✔️
Chi-Square	-	-	✔️	-	✔️	-	-

Test for 10 000 generated numbers	G	J	B	D	P	W	N
Series test	✔️	✔️	❌	✔️	❌	✔️	✔️
Chi-Square	-	-	✔️	-	✔️	-	-

Test for 50 000 generated numbers	G	J	B	D	P	W	N
Series test	✔️	✔️	❌	✔️	❌	✔️	✔️
Chi-Square	-	-	✔️	-	✔️	-	-

Test for 110 000 generated numbers	G	J	B	D	P	W	N
Series test	✔️	✔️	❌	✔️	❌	✔️	✔️
Chi-Square	-	-	✔️	-	✔️	-	-