Alex Herbert created STATISTICS-40:
--------------------------------------

             Summary: Better probability(x0, x1) function to use the survival 
probability
                 Key: STATISTICS-40
                 URL: https://issues.apache.org/jira/browse/STATISTICS-40
             Project: Apache Commons Statistics
          Issue Type: Improvement
          Components: distribution
    Affects Versions: 1.0
            Reporter: Alex Herbert


There is a default implementation for the probability within a range computed 
using:
{noformat}
p(x0, x1) = cdf(x1) - cdf(x0)
{noformat}
When the CDF computes values close to 1.0 then this probability range may be 
more accurately evaluated using the survival function:
{noformat}
p(x0, x1) = sf(x0) - sf(x1)
{noformat}
The switch point would be if CDF(x0) is above 0.5 as accuracy may be lost 
verses a high-precision survival function. A solution is to cache the median 
for the distribution:
{code:java}
double probability(double x0, double x1) {
    if (x0 > x1) {
        throw new DistributionException(
            DistributionException.INVALID_RANGE_LOW_GT_HIGH, x0, x1);
    }
    double xm = getMedian();
    return x0 < xm ?
        cumulativeProbability(x1) - cumulativeProbability(x0) :
        survivalProbability(x0) - survivalProbability(x1);
 }{code}
The method can be placed in the Abstract class for the distribution which will 
compute and cache the median on the first invocation (using inverse CDF(0.5)). 
Implementations with a known median may choose to override this method to 
return the median directly.

Note: When there is no survival function this will reduce to the same as using 
the CDF:
{noformat}
sf = 1 - cdf

sf(x0) - sf(x1) = (1 - cdf(x0)) - (1 - cdf(x1))
                = cdf(x1) - cdf(x0)
{noformat}
There is no loss of precision with the default survival function since if 
cdf(x0) is > 0.5 then 1-cdf(x0) is exact and likewise for cdf(x1).



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

Reply via email to