a)
To be able to estimate the correlation coefficient, we look at the line of best fit. If all the points
lie on the line then we have a Perfect Correlation between the variables (Credit Score and APR).
The values for the correlation coefficient range between -1 (perfect negative correlation), and 1
(perfect positive correlation).
In the problem here, we can see that there are some outliers, but overall, we have a strong
negative correlation (the higher the FICO score, the lower the APR). we can therefore,
estimate r to be -0.9
b)
The slope is that rate of change in APR as a result of 1 point increase in the FICO score.
The intercept is the APR for a person with a FICO score of 0.
c)
Yes, declaring causation is reasonable. It is reasonable to say that someone can lower their APR
on a 36 month car loans by increasing their FICO credit score. This is because the negative
slope shows an inverse relationship between APR and FICO credit score, the higher the FICO
score, the lower the APR, and the lower the FICO score, the higher the APR.
d)
No, this is not a reasonable idea because the regression line is based on FICO scores in the range
of 550-750, hence using the regression equation to predict the APR for a FICO score of 400
might not be reasonable.
e)
APR=61.369-0.076 (FICO Credit score)
APR=61.369−0.076 ( 660 )
APR = 11.209