As a sustainable mode with numerous benefits, cycling offers promoting physical fitness and providing a cost-effective transport option. To promote cycling, cities worldwide are striving to increase cycling usage by improving infrastructure, and network connectivity. Understanding the magnitude of the influence of cycling infrastructure on cycling usage is vital. This paper hence aims to estimate the impact of cycling infrastructure improvement on cycling usage. To this end, a generalised spatial difference-in-differences (GS-DiD) fixed effect spatial lag model is utilised using data collected from bicycle counters in Lyon and Paris. The model allows controlling spatial and temporal unobserved heterogeneity. The results indicate that cycling infrastructure improvement has a positive and statistically significant effect on cycling usage. Additionally, the spatial coefficient was found to be negative and statistically significant, suggesting that an increase in cycling infrastructure usage in neighbouring areas leads to a decrease in infrastructure usage in the area being examined. These findings provide evidence of the positive impact of cycling infrastructure improvement on cycling usage, while also highlighting the importance of considering spatial spillover effects in the analysis.

中文翻译：

---

估计自行车基础设施改进对使用的影响：空间双重差分法


作为一种具有众多好处的可持续方式，骑自行车可以促进身体健康并提供具有成本效益的交通选择。为了促进自行车运动，世界各地的城市都在努力通过改善基础设施和网络连接来增加自行车的使用。了解自行车基础设施对自行车使用的影响程度至关重要。因此，本文旨在评估自行车基础设施改善对自行车使用的影响。为此，利用从里昂和巴黎的自行车计数器收集的数据，利用广义空间双重差分（GS-DiD）固定效应空间滞后模型。该模型允许控制空间和时间上未观察到的异质性。结果表明，自行车基础设施的改善对自行车的使用具有积极且具有统计显着性的影响。此外，空间系数为负且具有统计显着性，表明邻近地区自行车基础设施使用量的增加导致所检查区域基础设施使用量的减少。这些发现提供了自行车基础设施改善对自行车使用产生积极影响的证据，同时也强调了在分析中考虑空间溢出效应的重要性。