Midterm

Question 1

variance

Answer 1

standard deviation^2

Question 2

variance of the mean

Answer 2

standard dev^2 / n

Question 3

SE of the mean

Answer 3

s / sqrt (n) OR sqrt (Stddev^2 / n)

Question 4

SE of the mean difference

Answer 4

square root: (Sa^2/Na + Sb^2/Nb)

Question 5

95% confidence interval implies p value is ____ and want |t| ______

Answer 5

95% confidence interval implies p value is less than .05 and want |t| greater than 1.96

Question 6

t = x-bar - u / SE(x)

x-bar =
Ux=
SE =

Answer 6

x-bar = sample average
Ux= population parameter of interest
SE = standard error

Question 7

Omitted Variable Bias
Correl x1, x2: +
B2: +
Bias =

Answer 7

Bias = +

Question 8

Omitted Variable Bias
Correl x1, x2: -
B2: -
Bias =

Answer 8

Bias = +

Question 9

Omitted Variable Bias
Correl x1, x2: +
B2: -
Bias =

Answer 9

Bias = -

Question 10

Omitted Variable Bias
Correl x1, x2: -
B2: +
Bias =

Answer 10

Bias = -

Question 11

Bias +
B1 -
over/under estimating bias?

Answer 11

underestimating

If the bias leads our regression coefficient to be larger in absolute value than it should be (i.e. if it moves us away from zero), we say that we are overstating (or overestimating) the effect of X on Y. This happens when B1 and the bias have the same sign

Question 12

Bias +
B1 +
over/under estimating bias?

Answer 12

overestimating

If the bias leads our regression coefficient to be larger in absolute value than it should be (i.e. if it moves us away from zero), we say that we are overstating (or overestimating) the effect of X on Y. This happens when B1 and the bias have the same sign

Question 13

Bias -
B1 -
over/under estimating bias?

Answer 13

overestimating

If the bias leads our regression coefficient to be larger in absolute value than it should be (i.e. if it moves us away from zero), we say that we are overstating (or overestimating) the effect of X on Y. This happens when B1 and the bias have the same sign

Question 14

Bias -
B1 +
over/under estimating bias?

Answer 14

underestimating

If the bias leads our regression coefficient to be larger in absolute value than it should be (i.e. if it moves us away from zero), we say that we are overstating (or overestimating) the effect of X on Y. This happens when B1 and the bias have the same sign

Question 15

Key Assumption 1

Answer 15

Xi: E(Ui|Xi) = 0

The conditional distribution of Ui given Xi has a mean of 0.

All factors contained in Ui are unrelated to Xi.

Question 16

Key Assumption 2

Answer 16

(Xi, Yi) i = 1…..n. All are independent and identically distributed

No patterns in the data

Question 17

Key Assumption 3

Answer 17

X and u have 4 finite movement.

No large outliers.

Question 18

Key Assumption 4

Answer 18

no strong multicollinearity

Question 19

linear-log

Answer 19

A 1% change in X is associated with a change in Y of .01*B1

Question 20

log-linear

Answer 20

A 1 unit change in X is associated with a change in Y of 100%*B1

Question 21

log-log

Answer 21

A 1% change in X is associated with a change in Y of B1%

Question 22

odds

Answer 22

1 / 1-p

Question 23

homoskedasticity vs. heteroskadisticity

Answer 23

variance of u|X = x is constant

variance of u|X =x is a function of x

Question 24

logit(Pi)

Answer 24

ln (p/1-p) = nu

–> pi = 1 / 1+e^nu