Changes

*i.e.,The Least Squares method relies on taking partial derivatives with respect to the slope  

*i.e.,The Least Squares method relies on taking partial derivatives with respect to the slope  

and intercept which provides a solvable pair of equations called normal equations.

and intercept which provides a solvable pair of equations called normal equations.

*Suppose there are <math> n </math> data points {y\_i, x\_i}, where i = 1, 2, …, n.

*Suppose there are <math> n </math> data points {y_{i}, x_{i}}, where i = 1, 2, …, n.

*To find the equation of the regression line:<math> a=y(bar)-b.x(bar)</math>.

*To find the equation of the regression line:<math> a=y(bar)-b.x(bar)</math>.

*This equation will give a "best" fit for the data points.  

*This equation will give a "best" fit for the data points.  

*The "best" means least-squares method. Here b is the slope.

*The "best" means least-squares method. Here b is the slope.

*The slope is calculated by:<math> b=summation(i=1 to n)(x_i-x(bar))(y_i-y(bar))/ summation(i=1 to n)[(x_i-x(bar))]^2.  

*The slope is calculated by:<math> b=summation(i=1 to n)(x_{i}-x(\bar))(y_{i}-y(\bar))/ summation(i=1 to n)[(x_i-x(bar))]^2.  

*In this formula<math> x(bar)</math> and<math> y(bar)</math> are the sample means  AVERAGE of <math> x</math>  and <math> y </math>.  

*In this formula<math> x(bar)</math> and<math> y(bar)</math> are the sample means  AVERAGE of <math> x</math>  and <math> y </math>.  

*In <math>INTERCEPT(y,x)</math> , the arguments can be numbers, names, arrays, or references that contain numbers.

*In <math>INTERCEPT(y,x)</math> , the arguments can be numbers, names, arrays, or references that contain numbers.