Working Paper: NBER ID: w19314
Authors: James J. Heckman; Rodrigo Pinto
Abstract: This paper presents an econometric mediation analysis. It considers identification of production functions and the sources of output effects (treatment effects) from experimental interventions when some inputs are mismeasured and others are entirely omitted.
Keywords: econometric mediation analysis; production functions; treatment effects; unmeasured inputs; mismeasured inputs
JEL Codes: C21; C38; C43; D24
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| confounding by unobserved variables (C29) | misattribution of effects (D91) |
| treatment effects (C22) | outputs (Y10) |
| treatment (M53) | measured inputs (C67) |
| treatment (M53) | unmeasured inputs (D24) |
| measured inputs (C67) | outputs (Y10) |
| unmeasured inputs (D24) | outputs (Y10) |
| treatment (M53) | outputs (Y10) |