Statistics Applied to Law
Time: 1h30’. Answer five questions out of the following six (max 4 points for each correct answer).
1. Assume that the correlation coefficient between X and Y is equal to 0. Can we conclude
that these two variables are completely unrelated? Why?
2. Briefly explain what multicollinearity is and what problems it may cause in regressions.
3. State the main properties of the mean.
4. What happens in a regression model when some relevant variable is omitted?
5. What does the R-squared of a regression represent?
6. Which are the measures through which you can assess whether a distribution is normal?
Answer the following question (max 10 points)
7. Consider the following results of an OLS regression:
_cons 24.6218 3 3.02627 8.14 0.000 18.68141 30.56225
con_ignoranza -2.64955 7 2.252396 -1.18 0.240 -7.070903 1.771789
intermedio -7.32412 7 2.286782 -3.20 0.001 -11.81297 -2.835283
senza_ignoranza 7.82289 8 2.22646 3.51 0.000 3.452462 12.19333
tempo_scelta .095956 4 .0296904 3.23 0.001 .0376755 .1542373
scelta Coef . Std. Err. t P>|t| [95% Conf. Interval]
Robust
Root MSE = 22.558
R-squ ared = 0.0699
Prob > F = 0.0000
F( 4 , 795) = 15.32
Linear regression Numbe r of obs = 800
Explain:
a. Where we can read p-values and what these mean
b. Which coefficients are statistically significant
c. What the “robust” option entails
d. What the “Coef.” column contains.
