Wednesday, January 22, 2014
Friday, September 23, 2011
Tuesday, June 14, 2011
Qin and Zhang 2011
Road to Specialization in Agricultural Production:
Tales of 18 Natural Villages in China
-Sees crop specialization occur due to transaction costs
http://ageconsearch.umn.edu/bitstream/103882/4/Impact%20of%20Road%20Access%20on%20Agricultural%20Production%20in%20China%202011%20AAEA.pdf
Tales of 18 Natural Villages in China
-Sees crop specialization occur due to transaction costs
http://ageconsearch.umn.edu/bitstream/103882/4/Impact%20of%20Road%20Access%20on%20Agricultural%20Production%20in%20China%202011%20AAEA.pdf
agriculture roads and ecnomic development in Uganda: Gollin Rogerson 2010
Uses GE and size of quasi subsistence sector. Show highly sensitive to agric prod levels and transportation costs
http://cbey.research.yale.edu/uploads/Environmental%20Economics%20Seminar/Gollin_and_Rogerson_-_Agriculture_Roads_and_Economic_Development_in_Uganda,_3-19-10.pdf
http://cbey.research.yale.edu/uploads/Environmental%20Economics%20Seminar/Gollin_and_Rogerson_-_Agriculture_Roads_and_Economic_Development_in_Uganda,_3-19-10.pdf
Thursday, May 26, 2011
Enhancing Anti-Corruption Programming in the Europe and Eurasia Region
Throughout the former communist states of the Europe and Eurasia region, endemic corruption constrains economic growth and democratic development. To diagnose root causes of corruption, IRIS developed an in-depth assessment tool that builds on, complements and overcomes the programming limitations of macro surveys and indices generated by such institutions as Transparency International, the World Bank, Freedom House, Coalition 2000 and the South East Legal Development Initiative (SELDI). Three pilot projects in the Europe and Eurasia region tested, refined, and demonstrated the tool in two different sectors, business licensing and entry regulation and health care. In Romania and Russia, the pilot application of the assessment tool focused on business licensing, and in Bulgaria the test addressed related aspects of health care delivery by hospitals. Each application involved developing, field testing, and then implementing surveys of both public and private sector actors.
Monday, November 29, 2010
Propensity Score Matching (PSM) Reference
2 Assumptions for treatment to be strongly ignorable (Rosenbaum Rubin 1983) thereby making PSM appropirate:
CIA (Conditional Independence Assump.): Conditional on observables (Xs) outcomes (Ys) are independent of treatment status (T=0 or 1).
Common Support: For each value of observables (Xs) there is a positive probability of being treated and untreated. In other words observe all levels of X in both groups (treated and untreated).
Pitfall--Curse of Dimensionality
The more observables the more difficult to find close matches across all Xs-hence need PS to gauge 'closeness'.
Matching Algorithms
Nearest Neighbor Matching
Radius Matching (calipers)
Kernel/local-linear matching (non-par that compares treated with w.avg. all all no treated; weighted by PS proximity.
Suggested test for robustness (Heinrich below)is to try all of them.(?)
Assumption and Spec Tests
For specification of the selection equations follow same rules as one would any other regression.
No great way of testing validity of CIA but can use to institutional knowledge to argue basis.
For common support one can test do an F-test (Hotelling test) on the joint X differences between treated and untreated.
Unobserved Heterogeneity and relaxing CIA
CIA can be somewhat relaxed if use DD on outcome comparisons. Of course this requires the assumption that if there was selection on non-observables the non-observables be *time invariant*. If that could strongly be argued then there is a strong case. (requires panel data too).
PSM References:
Heinrich, Maffioli, Vazquez, "A Primer for Applying Propensity-Score Matching", IDB Working paper 2010.
Angrist and Pirschke, "Mostly Harmless Econometrics", (Book) 2009.
Rosenbaum's "Observational Studies" (Book), 2002.
CIA (Conditional Independence Assump.): Conditional on observables (Xs) outcomes (Ys) are independent of treatment status (T=0 or 1).
Common Support: For each value of observables (Xs) there is a positive probability of being treated and untreated. In other words observe all levels of X in both groups (treated and untreated).
Pitfall--Curse of Dimensionality
The more observables the more difficult to find close matches across all Xs-hence need PS to gauge 'closeness'.
Matching Algorithms
Nearest Neighbor Matching
Radius Matching (calipers)
Kernel/local-linear matching (non-par that compares treated with w.avg. all all no treated; weighted by PS proximity.
Suggested test for robustness (Heinrich below)is to try all of them.(?)
Assumption and Spec Tests
For specification of the selection equations follow same rules as one would any other regression.
No great way of testing validity of CIA but can use to institutional knowledge to argue basis.
For common support one can test do an F-test (Hotelling test) on the joint X differences between treated and untreated.
Unobserved Heterogeneity and relaxing CIA
CIA can be somewhat relaxed if use DD on outcome comparisons. Of course this requires the assumption that if there was selection on non-observables the non-observables be *time invariant*. If that could strongly be argued then there is a strong case. (requires panel data too).
PSM References:
Heinrich, Maffioli, Vazquez, "A Primer for Applying Propensity-Score Matching", IDB Working paper 2010.
Angrist and Pirschke, "Mostly Harmless Econometrics", (Book) 2009.
Rosenbaum's "Observational Studies" (Book), 2002.
Subscribe to:
Posts (Atom)