posted 3 years ago

I had a similar problem a while back (computing probabilities for an arbitrary number of different dice (with arbitrary number of sides), and am interested in how you solved this

Life is full of choices. Sometimes you make the good ones, and sometimes you have to kill all the witnesses.

posted 3 years ago

I could be incorrect (it's been a while since I took probability), but I think each dice could be modeled by a discrete uniform probability mass function (PMF), which would be an array of probabilities. If you want to find the PMF of the sum of the dice rolls, then you have to convolve all the dice PMF's together. To get the probability that you get a particular dice roll sum, evaluate the PMF of the sum of the dice rolls at that value. Maybe these calculations could be optimized to avoid the usage of arrays if you're only looking for the probability of one particular sum at a time, but I'm not sure.

posted 3 years ago

You're correct.

But determining the (P)DF (or PMF) is not that easy. It makes for a nice problem though:

Given a Die class, with one member: number of sides, and a List of Dice,

give the PDF of this list. (i.e: for each possible outcome of a throw of all the dice

in the list, what is the chance of this outcome?).

Greetz,

Piet

But determining the (P)DF (or PMF) is not that easy. It makes for a nice problem though:

Given a Die class, with one member: number of sides, and a List of Dice,

give the PDF of this list. (i.e: for each possible outcome of a throw of all the dice

in the list, what is the chance of this outcome?).

Greetz,

Piet

posted 3 years ago

I fear I've misunderstood the problem then. What probability is it that you're trying to work out?

I was working on the premise that it had something to do with the "sum" (assuming it's actually possible) being either equalled or exceeded in the given number of throws, which is likely to be a fixed value and have very little to do with random numbers at all.

Or are you trying to get the probability returned by a randomised

Winston

Damien Sky wrote:Here's mine:

I fear I've misunderstood the problem then. What probability is it that you're trying to work out?

I was working on the premise that it had something to do with the "sum" (assuming it's actually possible) being either equalled or exceeded in the given number of throws, which is likely to be a fixed value and have very little to do with random numbers at all.

Or are you trying to get the probability returned by a randomised

*test*?

Winston

"Leadership is nature's way of removing morons from the productive flow" - Dogbert

Articles by Winston can be found here

Piet Souris

Master Rancher

Posts: 2045

75

posted 3 years ago

What Damien does is simply to approach the chances by using the so called 'Monte Carlo'

technique. Throw many times, and note how often you get a certain outcome. Nothing

wrong with that, it is just that we are used to getting exact outcomes in the case of

throwing dice, by simply counting how often we get a certain outcome.

This is not what we call (in Statistics) a 'randomized test'.

Greetz,

Piet

technique. Throw many times, and note how often you get a certain outcome. Nothing

wrong with that, it is just that we are used to getting exact outcomes in the case of

throwing dice, by simply counting how often we get a certain outcome.

This is not what we call (in Statistics) a 'randomized test'.

Greetz,

Piet

posted 3 years ago

Fair enough, but it's what

Winston

Piet Souris wrote:What Damien does is simply to approach the chances by using the so called 'Monte Carlo' technique. Throw many times, and note how often you get a certain outcome. Nothing wrong with that, it is just that we are used to getting exact outcomes in the case of throwing dice, by simply counting how often we get a certain outcome.

This is not what we call (in Statistics) a 'randomized test'.

Fair enough, but it's what

*I*meant . Now I understand.

Winston

Articles by Winston can be found here

With a little knowledge, a cast iron skillet is non-stick and lasts a lifetime. |