On the measurement of importance
When cost-benefit analysis is infeasible, most empirical papers use standardized beta coefficients, Shapley values, or partial R2 to demonstrate that their relationship of interest is economically important or economically significant. I show that these measures of importance are flawed, and therefo...
| Main Author: | |
|---|---|
| Format: | Working paper |
| Published: |
2019
|
| _version_ | 1797092978219548672 |
|---|---|
| author | Sterck, O |
| author_facet | Sterck, O |
| author_sort | Sterck, O |
| collection | OXFORD |
| description | When cost-benefit analysis is infeasible, most empirical papers use standardized beta coefficients, Shapley values, or partial R2 to demonstrate that their relationship of interest is economically important or economically significant. I show that these measures of importance are flawed, and therefore propose two new approaches to measure importance. While the ceteris paribus approach captures the independent effect of a regressor on the dependent variable, the non-ceteris paribus approach explicitly considers whether this effect is reinforcing or going against the effects of other correlated variables. I apply these complementary methods to assess the importance of the determinants of long-run growth. |
| first_indexed | 2024-03-07T03:53:44Z |
| format | Working paper |
| id | oxford-uuid:c22709e7-c123-404e-b6a4-5e2379079027 |
| institution | University of Oxford |
| last_indexed | 2024-03-07T03:53:44Z |
| publishDate | 2019 |
| record_format | dspace |
| spelling | oxford-uuid:c22709e7-c123-404e-b6a4-5e23790790272022-03-27T06:06:50ZOn the measurement of importanceWorking paperhttp://purl.org/coar/resource_type/c_8042uuid:c22709e7-c123-404e-b6a4-5e2379079027Symplectic Elements at Oxford2019Sterck, OWhen cost-benefit analysis is infeasible, most empirical papers use standardized beta coefficients, Shapley values, or partial R2 to demonstrate that their relationship of interest is economically important or economically significant. I show that these measures of importance are flawed, and therefore propose two new approaches to measure importance. While the ceteris paribus approach captures the independent effect of a regressor on the dependent variable, the non-ceteris paribus approach explicitly considers whether this effect is reinforcing or going against the effects of other correlated variables. I apply these complementary methods to assess the importance of the determinants of long-run growth. |
| spellingShingle | Sterck, O On the measurement of importance |
| title | On the measurement of importance |
| title_full | On the measurement of importance |
| title_fullStr | On the measurement of importance |
| title_full_unstemmed | On the measurement of importance |
| title_short | On the measurement of importance |
| title_sort | on the measurement of importance |
| work_keys_str_mv | AT stercko onthemeasurementofimportance |