Statistical Analysis

Statistical Analysis

Regression

Regression Analysis is defined as the technique to find out the relationship of variables. Commonly researches use this investigation technique to find the casual effect of variable on other. For instance, the impact of money supply, interest rate on inflation can be determined by regression analysis. Regression technique has now been more associated to the field of econometrics (Fry J et al, 2010). The analysis of linear regression allows us to study about description, explanation and prediction of variables(Hadi A & Chatterjee S,2013). In regression models variables are labeled according to their roles. The variable whose value changes due to other variables is called dependent variable. The variable which does not affect by other variables is called independent variable.

Types of Regression

There are two types of regression on the basis of number of variables used. They are as follow:

Bi-Variate Regression

As the name tells that “Bi” means “two” so this regression estimates the relationship only between two variables. One variable acts as dependent and the other one acts as independent. It is the simplest form of linear regression analysis. This is called simple linear regression because it explains the relationship of two variables. This analysis only use to examine the linear relationship.

Example

Where

Y = Dependent Variable

X= Independent Variable

= Intercept

= Slope of the variable

= Error Term

In the above equation two variables are used Y and X. Y is dependent variable and can be affected by the change in value of X while X is independent variable whose value cannot be change by Y. Slope i.e. in the equation tells us that how much change is caused in one variable due to change in other variable. It also shows the degree of steepness. is the intercept which defines the place on the Y axis bu which straight line passes. It shows the minimum level when the value of X=0 but Y will show some value. The last term of the equation that is refers to the error term of the equation. It represents the part of Y score that cannot be calculated for by its methodical relationship with values of X.

Multi-Variate Regression

There is more than one independent variable in multi-variate regression. Following is an example of multi-variate regression equation.

Where

Inflation = Dependent Variable

MS= Independent Variable

Int= Independent Variable

=Intercept

And = Slopes of the variable

= Error Term

In the above equation inflation is the dependent variable where as MS (Money Supply) and Int(Interest rate ) are independent variables.

Correlation

Correlation refers the relationship between two variables. It defines the association between the two variables that whether the variables are correlated with each other or not. It tells the quantitative value by which two variables go together. There are three forms of correlation:

Positive correlation

If the value of one variable(Y) changes with the same proportion as the other variable changes(X) means increases with the increase of Y and decreases with the decrease in the value of X then this correlation is called positive correlation.

Negative correlation

When the value of X decreases with the ...