Regression Analysis is a way of statistics to predict and forecast. It includes all techniques of modeling and analyzing several variables, in case of we have focus on the relationship between one or more independent variables and a dependent variable. Usually regression analysis helps to understand the changes in typical values of the dependent variables while anyone of the independent variable varies and other independent variables are fixed. It is also used to estimate the conditional expectations of the dependent and independent variables. Likewise, the independents variable’s estimation target is known as regression function and it is also used to define characterize the variation of the dependent variables which can be described with the help of a probability distribution.

Method of least squares was the earliest form of regression which was discovered by Legendre in 1805 and Gauss in 1809. Both used the method to solve the problem of determination of astronomical observations as the orbits of bodies around the Sun. Later in 1821, Gauss published his work as Gauss-Markov theorem. Regression analysis got its name from Francis Galton who used it in observations of a biological phenomenon and he used it to solve the problems of biology. But other researchers helped regression analysis to be more useful and applicable in other streams of education.

Regression analysis model uses following variables:

The unknown parameters (shown as B)

The independent variable (known as X)

The dependent variable (known as Y)

Besides this, regression analysis encircles various topics of statistical analysis including **statistical assumption, Linear regression, Underlying assumptions, General linear model , Regression diagnostics, Regression with “limited dependent” variables, Interpolation and extrapolation, Nonlinear regression, Power and sample size calculations** and so on.