Class BoundsOnRatiosInSampledSets

• public final class BoundsOnRatiosInSampledSets
extends Object
This class is used to compute the bounds on the estimate of the ratio |B| / |A|, where:
• |A| is the unknown size of a set A of unique identifiers.
• |B| is the unknown size of a subset B of A.
• a = |SA| is the observed size of a sample of A that was obtained by Bernoulli sampling with a known inclusion probability f.
• b = |SA ∩ B| is the observed size of a subset of SA.
Author:
Kevin Lang
• Method Detail

• getLowerBoundForBoverA

public static double getLowerBoundForBoverA(long a,
long b,
double f)
Return the approximate lower bound based on a 95% confidence interval
Parameters:
f - the inclusion probability used to produce the set with size a and should generally be less than 0.5. Above this value, the results not be reliable. When f = 1.0 this returns the estimate.
Returns:
the approximate upper bound
• getUpperBoundForBoverA

public static double getUpperBoundForBoverA(long a,
long b,
double f)
Return the approximate upper bound based on a 95% confidence interval
Parameters:
f - the inclusion probability used to produce the set with size a.
Returns:
the approximate lower bound
• getEstimateOfBoverA

public static double getEstimateOfBoverA(long a,
long b)
Return the estimate of b over a
Parameters:
Returns:
the estimate of b over a
• getEstimateOfA

public static double getEstimateOfA(long a,
double f)
Return the estimate of A. See class javadoc.
Parameters:
f - the inclusion probability used to produce the set with size a.
Returns:
the approximate lower bound
• getEstimateOfB

public static double getEstimateOfB(long b,
double f)
Return the estimate of B. See class javadoc.
Parameters: