Skip to main content
. 2020 Feb 7;15(2):e0228034. doi: 10.1371/journal.pone.0228034

Fig 1. Common Support for entire sample.

Fig 1

The final propensity score matching drops the observation that are off-support. The propensity scores were obtained from Stata command psmatch2 which were utilised in the estimation of the following PSM-DID model. However, it is worth noting that PSM reduces but does not eliminate the bias generated by unobservable confounding factors.