io.improbable.keanu.tensor

Class BivariateDataStatisticsCalculator

java.lang.Object

io.improbable.keanu.tensor.BivariateDataStatisticsCalculator

public class BivariateDataStatisticsCalculator
extends java.lang.Object

Calculates common statistics of a 2-dimensional data set Treats xData and yData as a series of points in the data set - their shapes must match.

Constructor Summary

Constructors

Constructor and Description
BivariateDataStatisticsCalculator(DoubleTensor xData, DoubleTensor yData)

Method Summary

Modifier and Type	Method and Description
double	estimatedGradient() Calculate the estimate of the gradient The regression coefficients (gradient and intercept) can be treated as random variables and estimated from the data.
double	estimatedIntercept() Calculate the estimate of the intercept The regression coefficients (gradient and intercept) can be treated as random variables and estimated from the data.
double	meanSquaredError() Calculate the MSE (mean squared error) which is an unbiased estimate of the variance of Y
long	size()
double	standardErrorForGradient() Calculate the standard error, which is the unbiased estimate of the standard deviation of the gradient From https://www2.isye.gatech.edu/~yxie77/isye2028/lecture12.pdf
double	standardErrorForIntercept() Calculate the standard error, which is the unbiased estimate of the standard deviation of the intercept From https://www2.isye.gatech.edu/~yxie77/isye2028/lecture12.pdf
double	xMean()
double	yMean()

Methods inherited from class java.lang.Object

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

Constructor Detail

BivariateDataStatisticsCalculator

public BivariateDataStatisticsCalculator(DoubleTensor xData,
                                         DoubleTensor yData)

Parameters:

xData - the data set for the predictor variable X

yData - the data set for the dependent variable Y

Method Detail

size

public long size()

Returns:

the size, i.e. number of points in the data set

xMean

public double xMean()

Returns:

The mean X value from the data set

yMean

public double yMean()

Returns:

The mean Y value from the data set

estimatedGradient

public double estimatedGradient()

Calculate the estimate of the gradient The regression coefficients (gradient and intercept) can be treated as random variables and estimated from the data. From https://www2.isye.gatech.edu/~yxie77/isye2028/lecture12.pdf

Returns:

the estimate of the gradient

estimatedIntercept

public double estimatedIntercept()

Calculate the estimate of the intercept The regression coefficients (gradient and intercept) can be treated as random variables and estimated from the data. From https://www2.isye.gatech.edu/~yxie77/isye2028/lecture12.pdf

Returns:

the estimate of the intercept

meanSquaredError

public double meanSquaredError()

Calculate the MSE (mean squared error) which is an unbiased estimate of the variance of Y

Returns:

The MSE

standardErrorForGradient

public double standardErrorForGradient()

Calculate the standard error, which is the unbiased estimate of the standard deviation of the gradient From https://www2.isye.gatech.edu/~yxie77/isye2028/lecture12.pdf

Returns:

The standard error

standardErrorForIntercept

public double standardErrorForIntercept()

Calculate the standard error, which is the unbiased estimate of the standard deviation of the intercept From https://www2.isye.gatech.edu/~yxie77/isye2028/lecture12.pdf

Returns:

The standard error