net.sf.jagg
Class LinearRegressionAggregator

java.lang.Object
  net.sf.jagg.Aggregator
      net.sf.jagg.TwoPropAggregator
          net.sf.jagg.LinearRegressionAggregator

public class LinearRegressionAggregator
extends TwoPropAggregator

This class represents the "linear regression" aggregator over two sets of numeric values. Many values can be returned by this Aggregator, all encapsulated in the object LinearRegressionStats.

Since:

0.1.0

Author:

Randy Gettman

See Also:

LinearRegressionStats

Field Summary

Fields inherited from class net.sf.jagg.Aggregator

PROP_SELF

Constructor Summary

LinearRegressionAggregator(java.lang.String properties)
Constructs a LinearRegressionAggregator on the specified properties, in the format: property, property2.

LinearRegressionAggregator(java.lang.String property, java.lang.String property2)
Constructs a LinearRegressionAggregator that operates on the specified properties.

Method Summary

void	init() Initialize the internal aggregators: A CovarianceAggregator, 2 VarianceAggregators (one each for both properties), and 2 AvgAggregators (one each for both properties).
void	iterate(java.lang.Object value) If both property values are non-null, then iterate the internal aggregators and increment the count.
void	merge(Aggregator agg) Merge the given Aggregator into this one.
LinearRegressionAggregator	replicate() Returns an uninitialized copy of this Aggregator object, with the same property(ies) to analyze.
LinearRegressionStats	terminate() Return a LinearRegressionStats, with the following calculations: slope = cov(prop1, prop2) / var(prop2) intercept = avg(prop1) - slope * avg(prop2) count = number of pairs where both elements were non-null, and thus counted in these calculations.

Methods inherited from class net.sf.jagg.TwoPropAggregator

getProperty2, setProperty, toString

Methods inherited from class net.sf.jagg.Aggregator

equals, getAggregator, getAggregator, getProperty, getValueFromProperty, hashCode, isInUse, setInUse, terminateDoubleDouble

Methods inherited from class java.lang.Object

clone, finalize, getClass, notify, notifyAll, wait, wait, wait

Constructor Detail

LinearRegressionAggregator

public LinearRegressionAggregator(java.lang.String properties)

Constructs a LinearRegressionAggregator on the specified properties, in the format: property, property2.

Parameters:

properties - A specification string in the format: property, property2.

LinearRegressionAggregator

public LinearRegressionAggregator(java.lang.String property,
                                  java.lang.String property2)

Constructs a LinearRegressionAggregator that operates on the specified properties.

Parameters:

property - Calculate linear regression statistics of this property with the other.

property2 - Calculate linear regression statistics of this property with the other.

Method Detail

replicate

public LinearRegressionAggregator replicate()

Returns an uninitialized copy of this Aggregator object, with the same property(ies) to analyze.

Specified by:

replicate in class Aggregator

Returns:

An uninitialized copy of this Aggregator object.

init

public void init()

Initialize the internal aggregators: A CovarianceAggregator, 2 VarianceAggregators (one each for both properties), and 2 AvgAggregators (one each for both properties). Initialize a count to zero.

Specified by:

init in class Aggregator

See Also:

CovariancePopAggregator, VariancePopAggregator, AvgAggregator

iterate

public void iterate(java.lang.Object value)

If both property values are non-null, then iterate the internal aggregators and increment the count.

Specified by:

iterate in class Aggregator

Parameters:

value - The value to aggregate.

See Also:

Aggregator.getValueFromProperty(java.lang.Object, java.lang.String)

merge

public void merge(Aggregator agg)

Merge the given Aggregator into this one. Add the internal counts.

Specified by:

merge in class Aggregator

Parameters:

agg - The Aggregator to merge into this one.

terminate

public LinearRegressionStats terminate()

Return a LinearRegressionStats, with the following calculations:

• slope = cov(prop1, prop2) / var(prop2)
• intercept = avg(prop1) - slope * avg(prop2)
• count = number of pairs where both elements were non-null, and thus counted in these calculations.
• rSquared = if (var(prop2)) is 0, then NaN,
  else if (var(prop1)) is 0, then 1,
  else correlation(prop1, prop2) squared.
• correlation = covariancePop(prop1, prop2) / Math.sqrt(variancePop(prop1) * variancePop(prop2))
• avg1 = avg(prop1)
• avg2 = avg(prop2)

Specified by:

terminate in class Aggregator

Returns:

A LinearRegressionStats.