Professional Documents
Culture Documents
Assignment of Fundamentals of Econometrics, On Multiple Regression
Assignment of Fundamentals of Econometrics, On Multiple Regression
Assignment of Fundamentals of Econometrics, On Multiple Regression
Multiple Regression.
Illustrations
In the below example I have done a multiple regression analysis, where my
dependent variable is profit of the company, and my independent variable are
administration spending, R&D spending and marketing spending.
The data is in the link below:
https://www.dropbox.com/s/z2ue4a1ogefcuo3/50_Startups.csv?dl=0
After doing this regression analysis we will find out by what proportion the
dependent variable i.e. profits of the company changes if one unit change in of
each of the independent variables (i.e. admin spending, R&D spending and
marketing spending) is made.
We will also know the interpretation of various symbols like x intercept (a), slope
(b), p value, R squared, and adjusted R squared.
SUMMARY OUTPUT
Regression Statistics
Multiple R 0.97110574
R Square 0.94304636
Adjusted R Square
0.93802104
Standard Error 7508.517
Observations 38
ANOVA
df SS MS F Significance F
Regression 3 3.1739E+10 1.058E+10 187.658912 3.228E-21
Residual 34 1916846135 56377827.5
Total 37 3.3656E+10
Coefficients Standard Error t Stat P-value Lower 95% Upper 95% Lower 95.0%Upper 95.0%
Intercept 56651.6482 7368.1816 7.68868783 6.1194E-09 41677.7016 71625.5948 41677.7016 71625.5948
R&D Spend 0.76930441 0.04740394 16.2287033 1.4025E-17 0.67296802 0.8656408 0.67296802 0.8656408
-0.0479275 0.05047484 -0.9495332 0.34905055 -0.1505048 0.05464969 -0.1505048 0.05464969
0.02666947 0.01560237 1.709322 0.09650525 -0.0050384 0.0583773 -0.0050384 0.0583773
Interpretation of X intercept and slopes in the regression model:
From the above analysis we can write the regression model as:
ŷ=56651 + 0.769 X1 -0.0479 X2 + 0.02667
where,
ŷ=Profit
X1= Administration spending
X2= R&D spending
X3= Marketing spending
Here we have x intercept which is equals to 56651. And 0.769, -0.0479, and
0.02667 as the slopes of each independent variables X1, X2 and X3 respectively.
As per the equation, if we increase administration spending by 1 value the profits
of the company will increase by only 0.769, keeping all other constant.
Similarly, if we increase R&D spending by 1 value then the profits of the company
will decrease by 0.0479 citers paribus.
And if we increase marketing spending by 1 rupee the profits of the company will
increase by 0.02667 rupees citers paribus.
Interpretation of P values:
Now focusing on the p values, any p value that shows a value greater than 0.5
(which is our level of significance) means that it does not have any strong
relationship with the Y value (i.e. profit) that we are trying to predict. statistically
not significant to reject the null hypothesis.
Here we do not have such p values whose value is greater than 0.5 hence in all
these independent variables we reject the null hypothesis which says that profit is
not affected by administration spending, R&D spending, and marketing spending
respectively; and accept the alternative which says that profits do get affected by
those spending.