Addons/stats/base/distribution

From J Wiki
Jump to navigation Jump to search
User Guide | Installation | Development | Categories | Git | Build Log

stats/base/distribution  - statistical distributions

binomialdist v discrete values for binomial distribution
binomialprob v probability of success in binomial distribution
chisqcdf v chi squared CDF
normalcdf v normal distribution (0,1) CDF
normalprob v probability of success in normal distribution
poissondist v discrete values for poisson distribution
poissonprob v probability of success in poisson distribution

binomialdist

binomialdist (v) discrete values for binomial distribution

y has 2 elements p,n:
 0  =  probability of success in one trial
 1  =  number of trials

E=n*p, var=n*p*-.p

e.g.  binomialdist 0.25 5  ...distribution of results from 0 to 5

binomialprob

binomialprob (v) probability of success in binomial distribution

y has 3 or 4 elements:
  0  =  probability of success in one trial
  1  =  number of trials
  2  =  minimum number of successes
 {3} =  maximum number of successes

e.g.  binomialprob 0.25 100 30 40    ...probability that there are
between 30 and 40 successes in 100 trials, where probability = 0.25

chisqcdf

chisqcdf (v) chi squared CDF

chi squared cumulative distribution function
see wiki Essays/Chi_Squared_CDF

normalcdf

normalcdf (v) normal distribution (0,1) CDF

normal(0,1) cumulative distribution function
see wiki Essays/Normal_CDF

the following are equivalent, but normalcdf more accurate:
  normalcdf X
  1 - normalprob 0 1, X

normalprob

normalprob (v) probability of success in normal distribution

y has 3 or 4 elements:
  0  =  mean of distribution
  1  =  standard deviation
  2  =  minimum result
 {3} =  maximum result

e.g.  normalprob 0 1 2    ...probability that result will be
greater than 2, where mean = 0, standard deviation 1

Result is rounded to 7 decimal places
Algorithm has accuracy ~ 7.5*1e_8

poissondist

poissondist (v) discrete values for poisson distribution

y has 2 elements:
  0 = mean of distribution
  1 = maximum value to show

e.g.  poissondist 2 10
      = list of probabilities of values from 0 to 10
        in poisson distribution of mean 2

poissonprob

poissonprob (v) probability of success in poisson distribution

y has 2 or 3 elements:
 0   =  mean of distribution
 1   =  minimum number of successes
 {2} =  maximum number of successes

e.g.
  poissonprob 5 4 7    ...probability that result will be
        between 4 and 7 inclusive, where mean result is 5.
  poissonprob 5 6 6    ...probability that result will be
        exactly 6, where mean result is 5.