Apache Commons Math 3.6.1 · org.apache.commons.math3.stat.regression.RegressionResults

Field Summary

Modifier and Type	Field and Description
private static final int	ADJRSQ_IDX INDEX of Adjusted R Squared
private final boolean	containsConstant boolean flag indicator of whether a constant was included
private final double[]	globalFitInfo array storing global results, SSE, MSE, RSQ, adjRSQ
private final boolean	isSymmetricVCD boolean flag for variance covariance matrix in symm compressed storage
private static final int	MSE_IDX INDEX of Mean Squared Error
private final long	nobs number of observations on which results are based
private final double[]	parameters regression slope parameters
private final int	rank rank of the solution
private static final int	RSQ_IDX INDEX of R-Squared of regression
private static final long	serialVersionUID UID
private static final int	SSE_IDX INDEX of Sum of Squared Errors
private static final int	SST_IDX INDEX of Sum of Squares of Model
private final double[][]	varCovData variance covariance matrix of parameters

Constructor Summary

Access	Constructor and Description
private	RegressionResults() Set the default constructor to private access to prevent inadvertent instantiation
public	RegressionResults(final double[] a double array with the regression slope estimates parameters, final double[][] the variance covariance matrix, stored either in a square matrix or as a compressed varcov, final boolean a flag which denotes that the variance covariance matrix is in symmetric compressed format isSymmetricCompressed, final long the number of observations of the regression estimation nobs, final int the number of independent variables in the regression rank, final double the sum of the independent variable sumy, final double the sum of the squared independent variable sumysq, final double sum of squared errors sse, final boolean true model has constant, false model does not have constant containsConstant, final boolean if true a deep copy of all input data is made, if false only references are copied and the RegressionResults become mutable copyData) Constructor for Regression Results.

Access

Constructor and Description

private

RegressionResults()

Set the default constructor to private access to prevent inadvertent instantiation

public

RegressionResults(final double[]

a double array with the regression slope estimates

parameters, final double[][]

the variance covariance matrix, stored either in a square matrix or as a compressed

varcov, final boolean

a flag which denotes that the variance covariance matrix is in symmetric compressed format

isSymmetricCompressed, final long

the number of observations of the regression estimation

nobs, final int

the number of independent variables in the regression

rank, final double

the sum of the independent variable

sumy, final double

the sum of the squared independent variable

sumysq, final double

sum of squared errors

sse, final boolean

true model has constant, false model does not have constant

containsConstant, final boolean

if true a deep copy of all input data is made, if false only references are copied and the RegressionResults become mutable

copyData)

Constructor for Regression Results.

Method Summary

Modifier and Type	Method and Description
public double	Returns: adjusted R-Squared statistic getAdjustedRSquared() Returns the adjusted R-squared statistic, defined by the formula R²_adj = 1 - [SSR (n - 1)] / [SSTO (n - p)]
public double	Returns: the covariance of the parameter estimates. getCovarianceOfParameters(int `i`th regression parameter. i, int `j`th regression parameter. j) Returns the covariance between regression parameters i and j.
public double	Returns: sum of squared errors associated with the regression model getErrorSumSquares() Returns the sum of squared errors (SSE) associated with the regression model.
public double	Returns: sum of squared deviations of y values getMeanSquareError() Returns the sum of squared errors divided by the degrees of freedom, usually abbreviated MSE.
public long	Returns: Number of observations, -1 if an error condition prevents estimation getN() Returns the number of observations added to the regression model.
public int	Returns: number of regressors, -1 if not estimated getNumberOfParameters() Returns the number of parameters estimated in the model.
public double	Returns: the parameters estimated for regressor at index. getParameterEstimate(int Index. index) Returns the parameter estimate for the regressor at the given index.
public double[]	Returns: array of parameter estimates, null if no estimation occurred getParameterEstimates() Returns a copy of the regression parameters estimates.
public double	Returns: sum of squared deviations of predicted y values getRegressionSumSquares() Returns the sum of squared deviations of the predicted y values about their mean (which equals the mean of y).
public double	Returns: r-square, a double in the interval [0, 1] getRSquared() Returns the coefficient of multiple determination, usually denoted r-square.
public double	Returns: the standard errors associated with parameters estimated at index. getStdErrorOfEstimate(int Index. index) Returns the standard error of the parameter estimate at index, usually denoted s(b_index).
public double[]	Returns: an array standard errors associated with parameters estimates, null if no estimation occurred getStdErrorOfEstimates() Returns the standard error of the parameter estimates, usually denoted s(b_i).
public double	Returns: sum of squared deviations of y values getTotalSumSquares() Returns the sum of squared deviations of the y values about their mean.
private double	Returns: the requested variance-covariance matrix entry getVcvElement(int first variable index i, int second variable index j) Gets the i-jth element of the variance-covariance matrix.
public boolean	Returns: true if the model has an intercept term hasIntercept() Returns true if the regression model has been computed including an intercept.

Inherited from java.lang.Object:: clone equals finalize getClass hashCode notify notifyAll toString wait wait wait

Field Detail

ADJRSQ_IDX	back to summary
private static final int ADJRSQ_IDX INDEX of Adjusted R Squared

containsConstant	back to summary
private final boolean containsConstant boolean flag indicator of whether a constant was included

globalFitInfo	back to summary
private final double[] globalFitInfo array storing global results, SSE, MSE, RSQ, adjRSQ

isSymmetricVCD	back to summary
private final boolean isSymmetricVCD boolean flag for variance covariance matrix in symm compressed storage

MSE_IDX	back to summary
private static final int MSE_IDX INDEX of Mean Squared Error

nobs	back to summary
private final long nobs number of observations on which results are based

parameters	back to summary
private final double[] parameters regression slope parameters

rank	back to summary
private final int rank rank of the solution Annotations @SuppressWarnings:unused

RSQ_IDX	back to summary
private static final int RSQ_IDX INDEX of R-Squared of regression

serialVersionUID	back to summary
private static final long serialVersionUID UID

SSE_IDX	back to summary
private static final int SSE_IDX INDEX of Sum of Squared Errors

SST_IDX	back to summary
private static final int SST_IDX INDEX of Sum of Squares of Model

varCovData	back to summary
private final double[][] varCovData variance covariance matrix of parameters

Constructor Detail

RegressionResults	back to summary
private RegressionResults() Set the default constructor to private access to prevent inadvertent instantiation Annotations @SuppressWarnings:unused

RegressionResults

back to summary

private RegressionResults()

Set the default constructor to private access to prevent inadvertent instantiation

Annotations

@SuppressWarnings:unused

RegressionResults	back to summary
public RegressionResults(final double[] parameters, final double[][] varcov, final boolean isSymmetricCompressed, final long nobs, final int rank, final double sumy, final double sumysq, final double sse, final boolean containsConstant, final boolean copyData) Constructor for Regression Results. Parameters parameters:double[] a double array with the regression slope estimates varcov:double[][] the variance covariance matrix, stored either in a square matrix or as a compressed isSymmetricCompressed:boolean a flag which denotes that the variance covariance matrix is in symmetric compressed format nobs:long the number of observations of the regression estimation rank:int the number of independent variables in the regression sumy:double the sum of the independent variable sumysq:double the sum of the squared independent variable sse:double sum of squared errors containsConstant:boolean true model has constant, false model does not have constant copyData:boolean if true a deep copy of all input data is made, if false only references are copied and the RegressionResults become mutable

RegressionResults

back to summary

public RegressionResults(final double[] parameters, final double[][] varcov, final boolean isSymmetricCompressed, final long nobs, final int rank, final double sumy, final double sumysq, final double sse, final boolean containsConstant, final boolean copyData)

Constructor for Regression Results.

Parameters

parameters:double[]: a double array with the regression slope estimates
varcov:double[][]: the variance covariance matrix, stored either in a square matrix or as a compressed
isSymmetricCompressed:boolean: a flag which denotes that the variance covariance matrix is in symmetric compressed format
nobs:long: the number of observations of the regression estimation
rank:int: the number of independent variables in the regression
sumy:double: the sum of the independent variable
sumysq:double: the sum of the squared independent variable
sse:double: sum of squared errors
containsConstant:boolean: true model has constant, false model does not have constant
copyData:boolean: if true a deep copy of all input data is made, if false only references are copied and the RegressionResults become mutable

Method Detail

getAdjustedRSquared back to summary

getAdjustedRSquared	back to summary
public double getAdjustedRSquared() Returns the adjusted R-squared statistic, defined by the formula R²_adj = 1 - [SSR (n - 1)] / [SSTO (n - p)] where SSR is the sum of squared residuals}, SSTO is the total sum of squares}, n is the number of observations and p is the number of parameters estimated (including the intercept). If the regression is estimated without an intercept term, what is returned is `1 - (1 - getRSquared() ) * (n / (n - p))` Returns:double adjusted R-Squared statistic

public double getAdjustedRSquared()

Returns the adjusted R-squared statistic, defined by the formula

R²_adj = 1 - [SSR (n - 1)] / [SSTO (n - p)]

where SSR is the sum of squared residuals}, SSTO is the total sum of squares}, n is the number of observations and p is the number of parameters estimated (including the intercept).

If the regression is estimated without an intercept term, what is returned is

 1 - (1 - getRSquared() ) * (n / (n - p))

Returns:double: adjusted R-Squared statistic

getCovarianceOfParameters back to summary

getCovarianceOfParameters	back to summary
public double getCovarianceOfParameters(int i, int j) throws OutOfRangeException Returns the covariance between regression parameters i and j. If there are problems with an ill conditioned design matrix then the covariance which involves redundant columns will be assigned `Double.NaN`. Parameters i:int `i`th regression parameter. j:int `j`th regression parameter. Returns:double the covariance of the parameter estimates. Exceptions OutOfRangeException: if `i` or `j` is not in the interval `[0, number of parameters)`.

public double getCovarianceOfParameters(int i, int j) throws OutOfRangeException

Returns the covariance between regression parameters i and j.

If there are problems with an ill conditioned design matrix then the covariance which involves redundant columns will be assigned Double.NaN.

Parameters

i:int: ith regression parameter.
j:int: jth regression parameter.

Returns:double

the covariance of the parameter estimates.

Exceptions

OutOfRangeException:: if i or j is not in the interval [0, number of parameters).

getErrorSumSquares back to summary

getErrorSumSquares	back to summary
public double getErrorSumSquares() Returns the sum of squared errors (SSE) associated with the regression model. The return value is constrained to be non-negative - i.e., if due to rounding errors the computational formula returns a negative result, 0 is returned. Preconditions: numberOfParameters data pairs must have been added before invoking this method. If this method is invoked before a model can be estimated, `Double,NaN` is returned. Returns:double sum of squared errors associated with the regression model

public double getErrorSumSquares()

Returns the sum of squared errors (SSE) associated with the regression model.

The return value is constrained to be non-negative - i.e., if due to rounding errors the computational formula returns a negative result, 0 is returned.

Preconditions:

numberOfParameters data pairs must have been added before invoking this method. If this method is invoked before a model can be estimated, Double,NaN is returned.

Returns:double: sum of squared errors associated with the regression model

getMeanSquareError back to summary

getMeanSquareError	back to summary
public double getMeanSquareError() Returns the sum of squared errors divided by the degrees of freedom, usually abbreviated MSE. If there are fewer than numberOfParameters + 1 data pairs in the model, or if there is no variation in `x`, this returns `Double.NaN`. Returns:double sum of squared deviations of y values

public double getMeanSquareError()

Returns the sum of squared errors divided by the degrees of freedom, usually abbreviated MSE.

If there are fewer than numberOfParameters + 1 data pairs in the model, or if there is no variation in x, this returns Double.NaN.

Returns:double: sum of squared deviations of y values

getN	back to summary
public long getN() Returns the number of observations added to the regression model. Returns:long Number of observations, -1 if an error condition prevents estimation

getN

back to summary

public long getN()

Returns the number of observations added to the regression model.

Returns:long: Number of observations, -1 if an error condition prevents estimation

getNumberOfParameters	back to summary
public int getNumberOfParameters() Returns the number of parameters estimated in the model. This is the maximum number of regressors, some techniques may drop redundant parameters Returns:int number of regressors, -1 if not estimated

getNumberOfParameters

back to summary

public int getNumberOfParameters()

Returns the number of parameters estimated in the model.

This is the maximum number of regressors, some techniques may drop redundant parameters

Returns:int: number of regressors, -1 if not estimated

getParameterEstimate back to summary

getParameterEstimate	back to summary
public double getParameterEstimate(int index) throws OutOfRangeException Returns the parameter estimate for the regressor at the given index. A redundant regressor will have its redundancy flag set, as well as a parameters estimated equal to `Double.NaN` Parameters index:int Index. Returns:double the parameters estimated for regressor at index. Exceptions OutOfRangeException: if `index` is not in the interval `[0, number of parameters)`.

public double getParameterEstimate(int index) throws OutOfRangeException

Returns the parameter estimate for the regressor at the given index.

A redundant regressor will have its redundancy flag set, as well as a parameters estimated equal to Double.NaN

Parameters

index:int: Index.

Returns:double

the parameters estimated for regressor at index.

Exceptions

OutOfRangeException:: if index is not in the interval [0, number of parameters).

getParameterEstimates back to summary

getParameterEstimates	back to summary
public double[] getParameterEstimates() Returns a copy of the regression parameters estimates. The parameter estimates are returned in the natural order of the data. A redundant regressor will have its redundancy flag set, as will a parameter estimate equal to `Double.NaN`. Returns:double[] array of parameter estimates, null if no estimation occurred

public double[] getParameterEstimates()

Returns a copy of the regression parameters estimates.

The parameter estimates are returned in the natural order of the data.

A redundant regressor will have its redundancy flag set, as will a parameter estimate equal to Double.NaN.

Returns:double[]: array of parameter estimates, null if no estimation occurred

getRegressionSumSquares back to summary

getRegressionSumSquares	back to summary
public double getRegressionSumSquares() Returns the sum of squared deviations of the predicted y values about their mean (which equals the mean of y). This is usually abbreviated SSR or SSM. It is defined as SSM here Preconditions: At least two observations (with at least two different x values) must have been added before invoking this method. If this method is invoked before a model can be estimated, `Double.NaN` is returned. Returns:double sum of squared deviations of predicted y values

public double getRegressionSumSquares()

Returns the sum of squared deviations of the predicted y values about their mean (which equals the mean of y).

This is usually abbreviated SSR or SSM. It is defined as SSM here

Preconditions:

At least two observations (with at least two different x values) must have been added before invoking this method. If this method is invoked before a model can be estimated, Double.NaN is returned.

Returns:double: sum of squared deviations of predicted y values

getRSquared back to summary

getRSquared	back to summary
public double getRSquared() Returns the coefficient of multiple determination, usually denoted r-square. Preconditions: At least numberOfParameters observations (with at least numberOfParameters different x values) must have been added before invoking this method. If this method is invoked before a model can be estimated, `Double,NaN` is returned. Returns:double r-square, a double in the interval [0, 1]

public double getRSquared()

Returns the coefficient of multiple determination, usually denoted r-square.

Preconditions:

At least numberOfParameters observations (with at least numberOfParameters different x values) must have been added before invoking this method. If this method is invoked before a model can be estimated, Double,NaN is returned.

Returns:double: r-square, a double in the interval [0, 1]

getStdErrorOfEstimate back to summary

getStdErrorOfEstimate	back to summary
public double getStdErrorOfEstimate(int index) throws OutOfRangeException Returns the standard error of the parameter estimate at index, usually denoted s(b_index). Parameters index:int Index. Returns:double the standard errors associated with parameters estimated at index. Exceptions OutOfRangeException: if `index` is not in the interval `[0, number of parameters)`.

public double getStdErrorOfEstimate(int index) throws OutOfRangeException

Returns the standard error of the parameter estimate at index, usually denoted s(b_index).

Parameters

index:int: Index.

Returns:double

the standard errors associated with parameters estimated at index.

Exceptions

OutOfRangeException:: if index is not in the interval [0, number of parameters).

getStdErrorOfEstimates back to summary

getStdErrorOfEstimates	back to summary
public double[] getStdErrorOfEstimates() Returns the standard error of the parameter estimates, usually denoted s(b_i). If there are problems with an ill conditioned design matrix then the regressor which is redundant will be assigned `Double.NaN`. Returns:double[] an array standard errors associated with parameters estimates, null if no estimation occurred

public double[] getStdErrorOfEstimates()

Returns the standard error of the parameter estimates, usually denoted s(b_i).

If there are problems with an ill conditioned design matrix then the regressor which is redundant will be assigned Double.NaN.

Returns:double[]: an array standard errors associated with parameters estimates, null if no estimation occurred

getTotalSumSquares back to summary

getTotalSumSquares	back to summary
public double getTotalSumSquares() Returns the sum of squared deviations of the y values about their mean. This is defined as SSTO here. If `n < 2`, this returns `Double.NaN`. Returns:double sum of squared deviations of y values

public double getTotalSumSquares()

Returns the sum of squared deviations of the y values about their mean.

This is defined as SSTO here.

If n < 2, this returns Double.NaN.

Returns:double: sum of squared deviations of y values

getVcvElement	back to summary
private double getVcvElement(int i, int j) Gets the i-jth element of the variance-covariance matrix. Parameters i:int first variable index j:int second variable index Returns:double the requested variance-covariance matrix entry

getVcvElement

back to summary

private double getVcvElement(int i, int j)

Gets the i-jth element of the variance-covariance matrix.

Parameters

i:int: first variable index
j:int: second variable index

Returns:double

the requested variance-covariance matrix entry

hasIntercept back to summary

hasIntercept	back to summary
public boolean hasIntercept() Returns true if the regression model has been computed including an intercept. In this case, the coefficient of the intercept is the first element of the `parameter estimates`. Returns:boolean true if the model has an intercept term

public boolean hasIntercept()

Returns true if the regression model has been computed including an intercept. In this case, the coefficient of the intercept is the first element of the parameter estimates.

Returns:boolean: true if the model has an intercept term

public Class RegressionResults

Field Summary

Constructor Summary

Method Summary

Field Detail

Constructor Detail

Method Detail