资源简介 (共14张PPT)TOOLS OF POSITIVE ANALYSISChapter 2Public Finance and Positive AnalysisOne goal of Public Finance is to estimate how government policy affects individuals’ behavior.2-*The Role of TheoryEconomic modelsProvide frameworks for thinking about the factors that might influence behavior.Generate hypotheses whose validity can be assessed through empirical work.Virtue of simplicity: reduces a problem to its essentials.Empirical analysis: Analysis based on observation and experience.Used to test hypotheses.2-*Causation vs. CorrelationConditions required for government action X to cause societal effect Y rather than just correlate (move together) with effect Y.X must precede YX and Y must be correlatedOther explanations for any observed correlation must be eliminatedThe importance of the distinction for policy.Example: There is a positive correlation between being married and wages…Does that mean government should enact a policy encouraging marriage as a way of increasing wages 2-*Empirical Work:Experimental StudiesExperimental (or randomized) study: subjects are randomly assigned to either a treatment group or control group.Treatment Group: Group of people who are subject to the intervention being studied.Control Group: Comparison group of people who are not subject to the intervention being studied.2-*Empirical Work:Experimental StudiesRandomization improves the chances that the control and treatment groups have similar characteristicsFocus can then be on possible causation between treatment and outcome.Randomization has a large potential to eliminate biased estimates.Biased estimate: conflates the true causal impact with the impact of outside factors.2-*Pitfalls of Experimental StudiesEthical issuesTechnical problemsResponse biasImpact of limited duration of experimentGeneralization of results to other populations, settings, and related treatmentsBlack box aspect of experiments2-*Empirical Work:Observational StudiesObservational study – empirical study relying on observed data not obtained from experimental studySources of observational dataSurveysAdministrative recordsGovernmental dataEconometrics: statistical techniques to establish and estimate causal relationships in absence of randomization.2-*Conducting Observational StudiesEstimating RelationshipsL = α0 + α1wn + α2X1 + … + αnXn + εDependent variableIndependent variablesParametersStochastic error termRegression analysisRegression line:Best fit through scatter of dataStandard errorIndicates reliability of equationwnLα0Intercept is α0Slope is α12-*Empirical Work:Types of Observational Study DataCross-sectional data: contains information on individual entities at a given point in time.Time-series data: contains information on a single entity at different points in time.Panel (longitudinal) data: combines features of both.Contains information on individual entities at different points in time.2-*Pitfalls of Observational StudiesData collected in non-experimental settingDifficult to ensure that the control group forms a valid “counterfactual”Counterfactual: the outcome for people in the treatment group had they not been treated.Specification issues – Not all variables that should be included are available in dataset and/or cannot be measured.2-*Empirical WorkQuasi-Experimental StudiesQuasi-experimental study (natural experiment) – observational study relying on circumstances outside researcher’s control to mimic random assignmentSuccessful Quasi-Experiments hinge on ensuring the treatment group assignment is random.Difference-in-Difference Quasi-Experiments.Instrumental Variables Quasi-ExperimentsRegression-Discontinuity Quasi-Experiments2-*Pitfalls of Quasi-Experimental StudiesAssignment to control and treatment groups may not be randomNot applicable to all research questionsGeneralization of results to other settings and treatments might not be possible.2-*Chapter 2 SummaryEconomic theory provides a framework for analyzing the causal relationship between government policy and individuals’ behavior.Empirical work tests hypothesis arising from economic theory to determine if it is consistent with real-world phenomena.Various methods for conducting empirical work exist.2-* 展开更多...... 收起↑ 资源预览