The Economic Geography of Road infrastructure in Pakistan: Exploratory Spatial Data Analysis (ESDA

Abstract

This paper presents an exploratory Spatial Data Analysis of road infrastructure using data of the district profiles of road infrastructure developed by the Urban Unit of the National Transport Research Centre (NTRC) during FY 2019-20. The indicators of the spatial autocorrelation (global and local Moran’s I) were estimated using the neighborhood definition of distance-based spatial weight matrix. Using inverse distance weights matrices at 3 and 5 bandwidths, results show significant positive spatial autocorrelation of road infrastructure at the national level of Pakistan, revealing that districts with a low density of road infrastructure are in the geographic vicinity of the districts of the less-dense road network. Likewise, the districts with more extensive road networks are neighboring the districts with large road networks. Furthermore, it is found that districts lying in the quadrant of the low-low road infrastructure are geographically located in the province of KPK, Sindh, and Baluchistan. While the districts lying in the high-high quadrants are located in the province of Punjab, Pakistan. In addition, this study found insignificant positive spatial autocorrelation of road infrastructure when analyzed at the provincial scale of Punjab, Sindh, KPK, and Baluchistan.