Summary: Microeconometrics

Study material generic cover image
  • This + 400k other summaries
  • A unique study and practice tool
  • Never study anything twice again
  • Get the grades you hope for
  • 100% sure, 100% understanding
Use this summary
Remember faster, study better. Scientifically proven.
Trustpilot Logo

Read the summary and the most important questions on Microeconometrics

  • 1

    This is a preview. There are 4 more flashcards available for chapter 03/01/2017
    Show more cards here

  • What are the possible positive correlations between income and health?

    1. Poeple with higher income get better medical care or can buy healthier food (I->H).
    2. People in ill health are often sick and get lower wages (H->I). 
    3. Poverty in childhood causes low income and poor health (P->I, P->H). 
    4. The finding is a chance result.
  • How are observable outcomes and potential outcomes related?

    Yi = Di*Y1i + (1-Di)*Y0i
  • Why is the average treatment effect not difference of observed outcomes?

    Treatment selection is often not independent of potential outcomes. People self-select into treatment.
  • When can we estimate counterfactuals?

    If there is some random variation in treatment that is not related with potential outcomes.
  • What are sources of exogenous variation in treatment?

    - Social experiments.
    - Natural experiments.
  • What is a solution of the self-selection problem?

    Treatment assignment is randomized across individuals. So treatment assignment is statistically independent of potential outcomes.
  • Which estimators can be used if random assignment holds?

    - The differences-in-means estimator.
    - OLS estimator.
  • Why can OLS regression be used if random assignment holds?

    Randomization ensures that Di is independently distributed from the unobserved factors in ui.
  • Treatment Evaluation

    This is a preview. There are 3 more flashcards available for chapter 04/01/2017
    Show more cards here

  • What is the role of covariates?

    Including covariates that have explanatory power for the outcome should be included because they reduce the standard error of the regression and thus increase the precision of the estimates.
  • Sources of bias (when experimenting in practice)

    1. Failure to randomize.
    2. Failure to follow treatment protocol (partial compliance).
    3. Experimental effects.

To read further, please click:

Read the full summary
This summary +380.000 other summaries A unique study tool A rehearsal system for this summary Studycoaching with videos
  • Higher grades + faster learning
  • Never study anything twice
  • 100% sure, 100% understanding
Discover Study Smart