Regression Discontinuity Designs with Nonclassical Measurement Error
| Project/Area Number |
15H06214
|
|---|
| Research Category |
Grant-in-Aid for Research Activity Start-up
|
| Allocation Type | Single-year Grants |
|---|
| Research Field |
Economic statistics
|
|---|
| Research Institution | Hitotsubashi University |
|---|
Principal Investigator |
YANAGI Takahide 一橋大学, 大学院経済学研究科, 講師 (30754832)
|
|---|
| Project Period (FY) |
2015-08-28 – 2017-03-31
|
| Project Status |
Completed (Fiscal Year 2016)
|
| Budget Amount *help |
¥2,470,000 (Direct Cost: ¥1,900,000、Indirect Cost: ¥570,000)
Fiscal Year 2016: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
Fiscal Year 2015: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
|
|---|
| Keywords | 経済統計学 / ミクロ計量経済学 / 政策評価 / 測定誤差 / ノンパラメトリック法 / 計量経済学 / ノンパラメトリック / 因果推論 |
|---|
| Outline of Final Research Achievements |
This project develops novel regression discontinuity (RD) inferences where the binary treatment and/or continuous assignment variable may contain measurement errors.
With a measurement error for the treatment, the standard RD estimator is inconsistent for the RD causal parameter since the measurement error for the binary variable is nonclassical by construction. To correct the problem, we propose a local linear generalized method of moments inference by utilizing the availability of an exogenous variable such as a covariate, instrument, or repeated measurement, and we derive its asymptotic properties. We then develop an identification analysis with a nonclassical measurement error for the assignment variable without additional information such as exogenous variables. Our analysis shows that, when there are units who accurately report their assignment values, the standard RD estimand may identify a meaningful causal parameter for such units.
|
Report
(3 results)
Research Products
(4 results)
