PP1: The Basics

Question 1

What are the properties of estimators?

Answer 1

1) Unbiasedness-unbiased if its expected value is equal to the trup parameter value
2) Efficiency-the unbaised estimator is efficient if it has the smallest variance

Question 2

Name and describe the 4 different data regression types.

Ex. cross-sectional

Answer 2

1)Cross-section-sample at one point in time
2)Time series-follow unit of observation over time
3)Pooled cross-sectional-samples over more than one point in time, but not the same unit(s) of observation
4)Panel-follow same units of observation over time

Question 3

What is the difference between percent and percentage point differece?

Answer 3

Percentage points is number changed, percent is number changed/original

Ex. increasing percent passed from 60 to 80 is a 20 per pt increase…

…and 33% increase (20/60)

Question 4

Define population

Answer 4

a collection of well-defined units

Question 5

T/F: Estimators are random variables

Answer 5

True

Question 6

What is a residual?

Answer 6

It’s what is “leftover” that is unexplained by the regressors

Question 7

What does OLS minimize and what does it do?

Answer 7

It minimizes the sum of squares of the difference beween the observed dependen variable to obain the “best fit” regression

Question 8

Interpret sqft in the following regression

price=-21.55+0.13sqft+12.49bdrms+13.08colonial

Where price is selling price in $1,000 and sqft is size of house in sqft

Answer 8

An increase of 1 square foot is predicted to incrase the house’s selling price by $130, holding other things fixed.

Question 9

T/F: R-squared helps us measure “goodness-of-fit”

Answer 9

True

Question 10

T/F: It’s a good idea to add as many explanatory variables as you can.

Answer 10

False: it may increase R^2 and make it look better. But are those variables economically sound?

Question 11

Interpret dist in the following regression.

log(price)=9.4+0.31log(dist)

Where price is sale price in $ and dist is distance from coal plant

Answer 11

When the distance increases by 1%, the house price increases by 0.31%

Question 12

Interpret bedrooms in the following regression

log(price)=10.2+0.084bedrooms+0.11baths+ 0.39log(area)

Where price is price of house and bedrooms is number of bedrooms

Answer 12

Holding everything else fixed, every additonal bedroom is predicted to increase the price by 8.4%

Question 13

Interpret expend in the following regression

math=-13.5-0.52lnchprg+7.7log(expend)-1.92log(enroll)

where math is % passing test and expend is school expenditure/student

Answer 13

Holding everything else fixed, increasing expenditures by 1% will increase the number of students passing mah by 0.077 percentage points.