Regression Project

REGRESSION PROJECT

I. DESCRIPTIVE ANALYSIS & DIFFERENCE IN MEANS

• OUTLIERS

• DESCRIPTIVE STATISTICS

With & Without Outlier

• ALPHA

• DIFFERENCE IN MEANS BY

o GENDER

o DEGREE

o POSITION

II. REGRESSION

• LOGIC

After examining relationships between each variable and salary, we are able to conclude that each variable will influence the salary paid to employees. However, the descriptive statistics only provide us a rough idea of the correlation between salary and individual factor. Further analysis thus would be needed to uncover relationships among variables corresponding to salary. Therefore, regression analysis, whose focus is on rationalizing the relationships between a dependent variable and one or more independent variables, has been chosen to justify the given sample data and to predict the trendline for the organization as a whole. By performing regression analysis, we will be able to discover the integration among independent variables such as degree, position, and tenure, and which variable(s) affect(s) the salary the most in order to reconcile the difference of salary paid between male and female.

• FINAL MODEL

• ELEMENTS (Meaning of Coefficients)

• STATISICAL ANALYSIS (R Squared, Significance, etc)

• Multicollinearity

Multicollinearity is designated to recognize high correlation between two or more independent variables in a multiple regression model. When two or more independent variables are highly correlated with each other, it means that one of the independent variables has insignificant impact on the dependent variable, or that the collinear variables contain the same information about the dependent variable. In this case, Age was highly correlated with Tenure. By...