Random

Type to generate random numbers. It uses Xorshift generators are among the fastest non-cryptographic random number generators. Xorshift random number generators are a class of pseudorandom number generators that was discovered by George Marsaglia (http://www.jstatsoft.org/v08/i14/paper). All the instances of Random along a given simulation have the same seed. The distribution can be inferred according to the selected arguments, as shown below.

Arguments	Default distribution
p	"bernoulli"
lambda	"poisson"
mean (or) sd	"normal"
min, max, step	"step"
min, max	"continuous"
(set of values)	"discrete"
(set of named values)	"categorical"

Arguments

alpha: Argument of beta distribution. The default value is 1.
beta: Argument of beta distribution. The default value is 1.
distrib: A string representing the statistical distribution to be used. See the table below.

Distrib	Description	Compulsory Arguments	Optional Arguments
"bernoulli"	A boolean distribution that returns true with probability p.	p	seed
"beta"	A family of continuous probability distributions defined on the interval [0, 1] parametrized by two positive shape parameters, denoted by alpha and beta, that appear as exponents of the random variable and control the shape of the distribution.		alpha, beta, seed
"categorical"	A distribution that has names associated to probabilities. Each name is an argument and has a value between zero and one, indicating the probability to be selected. The sum of all probabilities must be one.	...	seed
"continuous"	A continuous uniform distribition. It selects real numbers in a given interval.	max, min	seed
"discrete"	A discrete uniform distribition. Elements are described as a vector.	...	seed
"exponential"	Generate exponentialy distributed pseudo random numbers from a uniformly distributed number in the range [0,1]. For this purpose, it uses the Inverse Transform Samplig method		lambda, seed
"lognormal"	Generate log-normally distributed pseudo random numbers from a normaly distributed number in the range [0,1] using the Box-Muller (1978) method.		mean, sd, seed
"none"	No distribution. This is useful only when the modeler wants only to set seed.		seed
"normal"	Generate Normally (Gaussian) distributed pseudo random numbers from a uniformly distributed number in the range [0,1] using the Box-Muller (1978) method .		mean, sd, seed
"poisson"	Generate Poisson distributed pseudo random numbers from a uniformly distributed number in the range [0,1]. For this purpose, it uses the Inverse Transform Samplig method.		lambda, seed
"step"	A discrete uniform distribution whose values belong to a given [min, max] interval using step values.	max, min, step	seed
"weibull"	The Weibull distribution is a real valued distribution with two parameters a and b, producing values greater than or equals to zero.		k

k: The shape parameter for Weibull distribution. The default value is 1.
lambda: An argument of some distributions. It might be interpreted as mean or as scale, according to the given distribution. The default value is 1.
max: A number indicating the maximum value to be randomly selected.
mean: A number indicating the mean value. The default value is 1.
min: A number indicating the minimum value to be randomly selected.
p: A number between 0 and 1 representing a probability.
sd: A number indicating the standard deviation. The default value is 1.
seed: A number to generate the pseudo-random numbers. The default value is the current time of the system, which means that every simulation will use different random numbers. Choosing a seed in interesting when the modeler wants to have the same simulation outcomes despite using random numbers. It is a good programming practice to set the seed in the beginning of the simulation and only once.
step: The step where possible values are computed from minimum to maximum. When using this argument, min and max become mandatory.
...: Other values to build a categorical or discrete uniform distribution.

Attributes

Some attributes of Random have internal semantics. They can be used as read-only by the modeler.

distrib: The distribution of the Random object. All the other parameters of the distribution are also attributes.

Usage

random = Random()

bernoulli = Random{p = 0.4}
print(bernoulli:sample())

range = Random{min = 3, max = 7}
print(range:sample())

step = Random{min = 1, max = 9, step = 2}
print(step:sample())

gender = Random{male = 0.49, female = 0.51}
print(gender:sample())

age = Random{1, 2, 4, 8, 16, 32}
print(age:sample())

cover = Random{"pasture", "forest", "clearcut"}
print(cover:sample())

person = Agent{
    gender = Random{male = 0.49, female = 0.51},
    age = Random{mean = 20, sd = 2},
    contacts = Random{lambda = 5}
}

soc = Society{
    instance = person,
    quantity = 10
}

print(soc:gender().male)

Functions


integer	Return an integer random number.
number	Return a random real number.
reSeed	Set the seed to generate random numbers.
sample	Return a random element from the chosen distribution.

integer

Return an integer random number. It uses a discrete uniform distribution.

Arguments

#1: An integer number. If abscent, integer() will return zero or one. If it is the only argument, it will return a number between zero and this value.
#2: An integer number. When used, integer() will return a number between the first argument and the second, inclusive.

Usage

random = Random()

value = random:integer() -- 0 or 1
value = random:integer(10) -- from 0 to 10
value = random:integer(5, 10) -- from 5 to 10

number

Return a random real number. By default number() will return a value between zero and one.

Arguments

#1: A number. If it is the only argument used, it will return a number from zero to this value.
#2: A number. When used, number() will return a number between the first argument and the second.

Usage

random = Random()

value = random:number() -- between 0 and 1
value = random:number(10) -- between 0 and 10
value = random:number(5, 10) -- between 5 and 10

reSeed

Set the seed to generate random numbers. This seed will be used in new instances of Random. All Random objecs previously created will still use the previous seed.

Arguments

#1: An integer number with the new seed.

Usage

random = Random()

random:reSeed(12345)

sample

Return a random element from the chosen distribution.

Usage

random = Random{2, 3, 4, 6}

random:sample()