| 基于熵权法-BP神经网络的新质生产力测度及其对农业碳排放强度的影响机制 |
| 摘要点击 332 全文点击 15 投稿时间：2025-08-07 修订日期：2025-11-01 |
| 查看HTML全文
查看全文 查看/发表评论 下载PDF阅读器 |
| 中文关键词 新质生产力 农业碳排放 熵权法-BP神经网络 非线性 空间效应 |
| 英文关键词 new quality productive forces agricultural carbon emissions entropy weighting method-BP neural network nonlinearity spatial effects |
| DOI 10.13227/j.hjkx.202508069 |
|
| 中文摘要 |
| 新质生产力如何影响农业碳排放强度是值得研究的重要问题. 基于2012~2023年中国省域数据，运用BP神经网络优化熵权法对新质生产力进行系统性测度，并构建多维固定效应模型探究新质生产力对农业碳排放强度的非线性影响. 结果表明：①我国农业碳排放强度在2012~2023年间呈下降趋势，空间集聚态势逐步减弱；全国新质生产力发展水平整体呈上涨态势，但区域间差异显著. ②新质生产力对农业碳排放强度的影响呈显著的“U”型关系，且该效应在东北地区、主要粮食产区、非长江经济带地区以及长江经济带战略覆盖区尤为突出. ③新质生产力对农业机械化水平存在先抑制后促进的“U”型影响效应，而对农业信息化水平存在先促进后抑制的倒“U”型影响效应，二者在新质生产力影响农业碳排放强度的过程中均发挥了部分中介效应. ④新质生产力发展初期对农业碳排放强度的抑制作用会溢出至邻近地区，但在发展后期则会对邻近地区的农业碳排放强度产生促进作用. |
| 英文摘要 |
| The mechanism through which new quality productive forces （NQPF） influence agricultural carbon emission intensity constitutes a critical scientific question demanding in-depth investigation. Utilizing provincial panel data from China spanning 2012-2023， this study systematically measures NQPF by employing an entropy weighting method optimized via a BP neural network. A multidimensional fixed-effects model is constructed to empirically examine the complex nonlinear relationship between NQPF and agricultural carbon emission intensity. The results reveal that： ① China's agricultural carbon emission intensity showed a declining trend from 2012 to 2023， with a gradual weakening of spatial agglomeration. The overall development level of NQPF increased， but with significant regional disparities. ② NQPF exhibited a significant U-shaped relationship with agricultural carbon emission intensity， manifesting an inhibitory effect during initial development stages that transitioned to a promotional effect in later phases. This nonlinear pattern was particularly pronounced in Northeast China， major grain-producing regions， non-Yangtze River Economic Belt areas， and Yangtze River Economic Belt strategy-covered zones. ③ NQPF demonstrated a U-shaped association with agricultural mechanization levels but an inverted U-shaped relationship with agricultural informatization levels. Both factors exhibited significant partial mediating effects in the pathway through which NQPF influenced agricultural carbon emission intensity. ④ During early development stages， the inhibitory effect of NQPF on agricultural carbon emissions spilled over to neighboring regions， reducing their emission intensity. Conversely， in later stages， NQPF generated positive spatial spillovers that increased agricultural carbon emission intensity in adjacent areas. |
