13. Panel data

Question 1

What method should we use to carry out policy analysis using data with a time and a cross-sectional dimension?

Answer 1

Use differences in differences method

Question 2

How can an independently pooled cross-section be defined?

Answer 2

• Random sampling from a large population
• but not the same individuals or properties in each time period

Question 3

What does pooling data mean?

Answer 3

Combining the data sets together

Question 4

How can we measure changes in our data using dummies?

Answer 4

Using time dummies

Question 5

What does the Chow test measure?

Answer 5

Does your model stay the same across two different periods. It measures the level of structural change

Question 6

What does the constant mean with dummy variables?

Answer 6

It means the people who don’t fall into any of those dummies

Question 7

What do we have to assume to use the dif-in-dif model?

Answer 7

The parallel trends assumption

Question 8

What does the parallel trends assumption assume?

Answer 8

Difference in difference only works if the difference in outcomes between the two groups is not changed by other factors than the policy change (no differential trends)