Package com.compomics.util.math.statistics.linear_regression

Class LinearRegression

java.lang.Object

com.compomics.util.math.statistics.linear_regression.LinearRegression

public class LinearRegression
extends Object

Enum of the different implementations for a regression.

Author:

Marc Vaudel

Constructor Summary

Constructors

Constructor	Description
LinearRegression()	Empty default constructor

Method Summary

Modifier and Type	Method	Description
static RegressionStatistics	getRobustLinearRegression(ArrayList<Double> x, ArrayList<Double> y)	Returns a robust linear regression based on the median.
static RegressionStatistics	getSimpleLinearRegression(ArrayList<Double> x, ArrayList<Double> y)	Returns a simple linear regression.
static RegressionStatistics	getSimpleLinearRegressionOutlierRemoval(ArrayList<Double> x, ArrayList<Double> y, Double p)	Returns a simple linear regression performed after outlier removal.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Details

LinearRegression

public LinearRegression()

Empty default constructor

Method Details

getSimpleLinearRegression

public static RegressionStatistics getSimpleLinearRegression(ArrayList<Double> x,
 ArrayList<Double> y)

Returns a simple linear regression.

Parameters:

x - the x series

y - the y series

Returns:

a simple linear regression

getRobustLinearRegression

public static RegressionStatistics getRobustLinearRegression(ArrayList<Double> x,
 ArrayList<Double> y)

Returns a robust linear regression based on the median.

Parameters:

x - the x series

y - the y series

Returns:

a simple linear regression

getSimpleLinearRegressionOutlierRemoval

public static RegressionStatistics getSimpleLinearRegressionOutlierRemoval(ArrayList<Double> x,
 ArrayList<Double> y,
 Double p)

Returns a simple linear regression performed after outlier removal. If less than 100 points are available before or after filtering, a robust regression is used.

Parameters:

x - the x series

y - the y series

p - the probability for outlier exclusion, e.g. 0.95 for 95% confidence

Returns:

a simple linear regression