Advancing UN Comtrade for physical trade flow analysis: Addressing the issue of missing values

Trade contributes to the redistribution of resources among countries and regions. One of the most widely used data sources is the United Nations Commodity Trade Statistics Database (UN Comtrade). Nevertheless, data issues still limit its validity, trustworthiness, and use. A critical issue is the lack of commodity weight infor-mation. It relies heavily on data quality to determine the global market's suppliers and consumers. Thus, trade needs reliable methods for filling in the missing physical values. Using statistical approaches, we estimate missing physical values for commodities, countries/areas, and years. The impact of handled data on countries and commodities varies considerably; for example, South Africa's net weight rose by 117% and clocks' and watches' by 63% (HS0, 1988-2019), compared with their original data. The directions of net trade flows for 10594 records have been reversed. Finally, the bilateral asymmetry problem improved. Overall, this paper in-troduces a novel approach for improving data accuracy.

Automated summary

This paper introduces a novel approach for improving data accuracy in trade, where missing physical values are estimated using statistical methods, resulting in varying effects on countries and commodities during data processing.