This class provides an efficient online algorithm for calculating linear square line fit. The class also calculates the mean for the X's and Y's, and variance for the X's. These are byproducts of calculating the line fit. The online algorithm is detailed in the implementation notes.

Synopsis

Declared in <bdlsta_linefit.h>

class LineFit;

Enums

Name

Description

Unnamed enum

Status codes returned by statistical computations.

Member Functions

Name

Description

LineFit [constructor]

Create an empty LineFit object.

add

Add the specified (xValue, yValue) point to the data set.

count

Returns the number of elements in the data set.

fit

Calculate line fit coefficients Y = Alpha + Beta * X, and populate the specified alpha (intercept) and beta (slope). The behavior is undefined if 2 > count or all X's are identical.

fitIfValid

Calculate line fit coefficients Y = Alpha + Beta * X, and populate the specified alpha (intercept) and beta (slope). Return 0 on success, and non‐zero otherwise. The computations is unsuccessful if 2 > count or all X's are identical.

variance

Return the variance of the data set X's. The behavior is undefined unless 2 <= count.

varianceIfValid

Load into the specified result, the variance of the data set X's. Return 0 on success, and a non‐zero value otherwise. Specifically, e_INADEQUATE_DATA is returned if 2 > count.

xMean

Return the mean of the data set X's. The behavior is undefined unless 1 <= count.

xMeanIfValid

Load into the specified result, the mean of the data set X's. Return 0 on success, and a non‐zero value otherwise. Specifically, e_INADEQUATE_DATA is returned if 1 > count.

yMean

Return the mean of the data set Y's. The behavior is undefined unless 1 <= count.

yMeanIfValid

Load into the specified result, the mean of the data set Y's. Return 0 on success, and a non‐zero value otherwise. Specifically, e_INADEQUATE_DATA is returned if 1 > count.

Created with MrDocs