Abstract
Researchers often make use of linear regression models in order to assess the\r\nimpact of policies on target outcomes. In a correctly specified linear regression\r\nmodel, the marginal impact is simply measured by the linear regression coefficient.\r\nHowever, when dealing with both synchronic and diachronic spatial data, the\r\ninterpretation of the parameters is more complex because the effects of policies\r\nextend to the neighboring locations. Summary measures have been suggested in the\r\nliterature for the cross-sectional spatial linear regression models and spatial panel\r\ndatamodels. Inthis article,wecompare threeprocedures fortestingthesignificance\r\nofimpactmeasuresinthespatiallinearregressionmodels.Theseproceduresinclude\r\n(i) the estimating equation approach, (ii) the classical delta method, and (iii) the\r\nsimulationmethod.InaMonteCarlostudy,wecomparethefinitesampleproperties\r\nof these procedures.
Original language | English |
---|---|
Pages (from-to) | 1-36 |
Number of pages | 36 |
Journal | International Regional Science Review |
Volume | 2019 |
Issue number | 1 |
DOIs | |
Publication status | Published - 2019 |
All Science Journal Classification (ASJC) codes
- General Environmental Science
- General Social Sciences
Keywords
- MLE
- asymptotic approximation
- direct effects
- impactmeasures
- indirect effects
- inference
- spatialautoregressivemodels
- spatialeconometricmodels
- standard errors
- total effects