I have three variables HourlyWage YearsofSchooling and Work

I have three variables, HourlyWage, YearsofSchooling and Work Experience.

I define the variable lnHourlyWageStar as ln(HourlyWage/Geometric mean of Hourly Wage)

Now I regress lnHourlyWageStar on YearsofSchooling and WorkExperience

What\'s the interpertation of the respective coefficients? I understand the log-linear model, but I don\'t understand what happens when you divide it by the geometric mean.

If now I regress lnHourlyWage on YearsofSchooling and WorkExperience individually. I\'ll get different coefficients, why are they different?

The correlation between YearsofSchooling and WorkExperience is negative.

Thank you.

Solution

The regression line is of the form y =a+bx where b is the slope of the line

The coefficient of x or slope of the line interprets the rate of change of y when one unit of x is increased.

The coefficients for different variables are different becuase the rate of change on different coefficients are different.

The correlation between YearsofSchooling and WorkExperience is negative.

When correlation is negative we find that when one variable increases other variable decreases.

But there is an association though negative remains between them.