Table 4. The impact of climate change on agriculture input productivity.
| Variables | Model(1) | Model(2) | Model(3) |
|---|---|---|---|
| Land productivity | Labor productivity | Fertilizer productivity | |
| Agricultural methane | -5.80–08 (3.55–08) | -2.89–08 *** (5.77–09) | -.000018*** (4.57–06) |
| CO2 emission | -5.13–08*** (7.68–09) | -8.16–09 *** (1.25–09) | 1.46–06 (9.89–07) |
| GDP per capita income | 3.39–06 (2.35–07)*** | 4.83–07 *** (3.82–08) | .0003466 *** (.0000303) |
| constant | -.0013092 (.0008151) | .0005729 (.0001325) | 6.012633*** (.1052361) |
| Mean dependent var | 0.004 | 0.001 | 10125.080 |
| Overall R2 | 0.92 | 0.91 | 0.702 |
| x 2 | 243.763 | 243.100 | 1.328 |
| Number ofobservation | 595 | 595 | 594 |
| Prob > x2 **** | 0.000 | 0.000 | 0.723 |
| Breusch and Pagan Lagrangian multiplier test | |||
| Prob > chibar2 | 1.00 | 1.00 | 1.00 |
| Wooldridge test for autocorrelation in panel data | |||
| No Autocorrelation | F(1, 16) = 2.166 Prob > F = 0.1605 | F(1, 16) = 3.665 Prob > F = 0.0736 | F(1, 16) = 1.879 Prob > F = 0.1894 |
Source: own computation STATA 17
Note: in the above regression models results (***, **&*) are represented the 1%, 5% and 10% statistically level of significance. Diagnostic tests have confirmed no serial correlation and non-constant variance.