com.imsl.stat

Class LinearRegression.CaseStatistics

java.lang.Object

com.imsl.stat.LinearRegression.CaseStatistics

Enclosing class:

LinearRegression

public class LinearRegression.CaseStatistics
extends Object

Inner Class CaseStatistics allows for the computation of predicted values, confidence intervals, and diagnostics for detecting outliers and cases that greatly influence the fitted regression.

Constructor Summary

Constructors

Constructor and Description
CaseStatistics(double[] x, double y) Deprecated. The CaseStatistics constructors have been deprecated in favor of getter methods in LinearRegression.
CaseStatistics(double[] x, double y, double w) Deprecated. The CaseStatistics constructors have been deprecated in favor of getter methods in LinearRegression.
CaseStatistics(double[] x, double y, double w, int pred) Deprecated. The CaseStatistics constructors have been deprecated in favor of getter methods in LinearRegression.
CaseStatistics(double[] x, double y, int pred) Deprecated. The CaseStatistics constructors have been deprecated in favor of getter methods in LinearRegression.

Method Summary

Modifier and Type	Method and Description
double[]	getCaseStatistics() Deprecated. Recommend using specific getter methods.
double[]	getConfidenceInterval() Returns the Confidence Interval of the population mean for an observation.
double	getCooksDistance() Returns Cook's Distance for an observation.
double	getDFFITS() Returns DFFITS for an observation.
double	getJackknifeResidual() Returns the Jackknife Residual for an observation.
double	getLeverage() Returns the Leverage for an observation.
double	getObservedResponse() Returns the observed response for an observation.
double	getPredictedResponse() Returns the predicted response for an observation.
double[]	getPredictionInterval() Returns the Prediction Interval of the predicted response for an observation.
double	getResidual() Returns the Residual for an observation.
double	getStandardizedResidual() Returns the Standardized Residual for an observation.
void	setConLevelMean(double conpcm) Sets the confidence level for two-sided confidence intervals of the population mean.
void	setConLevelPred(double conpcp) Sets the confidence level for two-sided prediction intervals.

Methods inherited from class java.lang.Object

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

Constructor Detail

CaseStatistics

public CaseStatistics(double[] x,
                      double y)

Deprecated. The CaseStatistics constructors have been deprecated in favor of getter methods in LinearRegression.

Constructor for case statistics for an observation.

Parameters:

x - a double array containing the independent (explanatory) variables. Its length must be equal to the number of variables set in the LinearRegression constructor.

y - a double representing the dependent (response) variable

CaseStatistics

public CaseStatistics(double[] x,
                      double y,
                      double w)

Deprecated. The CaseStatistics constructors have been deprecated in favor of getter methods in LinearRegression.

Constructor for case statistics for an observation and a weight.

Parameters:

x - a double array containing the independent (explanatory) variables. Its length must be equal to the number of variables set in the constructor.

y - a double representing the dependent (response) variable

w - a double representing the weight

CaseStatistics

public CaseStatistics(double[] x,
                      double y,
                      double w,
                      int pred)

Deprecated. The CaseStatistics constructors have been deprecated in favor of getter methods in LinearRegression.

Constructor for case statistics for a new observation, weight, and future response count for the desired prediction interval.

Parameters:

x - a double array containing the independent (explanatory) variables. Its length must be equal to the number of variables set in the constructor.

y - a double representing the dependent (response) variable

w - a double representing the weight

pred - an int representing the number of future responses for which the prediction interval is desired on the average of the future responses

CaseStatistics

public CaseStatistics(double[] x,
                      double y,
                      int pred)

Deprecated. The CaseStatistics constructors have been deprecated in favor of getter methods in LinearRegression.

Constructor for case statistics for a new observation and future response count for the desired prediction interval.

Parameters:

x - a double array containing the independent (explanatory) variables. Its length must be equal to the number of variables set in the constructor.

y - a double representing the dependent (response) variable

pred - an int representing the number of future responses for which the prediction interval is desired on the average of the future responses.

Method Detail

getCaseStatistics

public double[] getCaseStatistics()

Deprecated. Recommend using specific getter methods.

Returns the case statistics for an observation.

Returns:

a double[12] array containing the case statistics. Elements 0 through 11 contain the following:

Index	Description
0	Observed response
1	Predicted response
2	Residual
3	Leverage
4	Standardized residual
5	Jackknife residual
6	Cook's distance
7	DFFITS
8,9	Confidence interval on the mean
10,11	Prediction interval

getObservedResponse

public double getObservedResponse()

Returns the observed response for an observation.

Returns:

a double containing the observed response for an observation

getPredictedResponse

public double getPredictedResponse()

Returns the predicted response for an observation.

Returns:

a double containing the predicted response for an observation

getResidual

public double getResidual()

Returns the Residual for an observation.

Returns:

a double containing the residual for an observation

getLeverage

public double getLeverage()

Returns the Leverage for an observation.

Returns:

a double containing the Leverage for an observation

getStandardizedResidual

public double getStandardizedResidual()

Returns the Standardized Residual for an observation.

Returns:

a double containing the Standardized Residual for an observation

getJackknifeResidual

public double getJackknifeResidual()

Returns the Jackknife Residual for an observation.

Returns:

a double containing the Jackknife Residual for an observation

getCooksDistance

public double getCooksDistance()

Returns Cook's Distance for an observation.

Returns:

a double containing Cook's Distance for an observation

getDFFITS

public double getDFFITS()

Returns DFFITS for an observation.

Returns:

a double containing the DFFITS value for an observation

getConfidenceInterval

public double[] getConfidenceInterval()

Returns the Confidence Interval of the population mean for an observation.

The width of the confidence interval depends on the confidence level set by LinearRegression.CaseStatistics.setConLevelMean(double).

Returns:

a double[2] array containing the Confidence Interval of the population mean at the given observation.

getPredictionInterval

public double[] getPredictionInterval()

Returns the Prediction Interval of the predicted response for an observation.

The width of the prediction interval depends on the confidence level set by LinearRegression.CaseStatistics.setConLevelPred(double).

Returns:

a double[2] array containing the Prediction Interval of the predicted response at the given observation

setConLevelPred

public void setConLevelPred(double conpcp)

Sets the confidence level for two-sided prediction intervals.

Parameters:

conpcp - a double used to set the confidence level for two-sided prediction intervals of a predicted response. Note that conpcp must be in the interval 0 to 1 inclusive. That is, it must be expressed as a probability.

Default: conpcp=.95.

setConLevelMean

public void setConLevelMean(double conpcm)

Sets the confidence level for two-sided confidence intervals of the population mean.

Parameters:

conpcm - a double used to set the confidence level for two-sided confidence intervals of the population mean. Note that conpcm must be in the interval 0 to 1 inclusive. That is, it must be expressed as a probability.

Default: conpcm=.95.