STOCHASTIC IMPUTATION OF MISSING PHYSICAL COMMODITY Y TRADE INFORMATION USING MONETARY TRADE DATA

摘要

International trade has conventionally been expressed in monetary values until therecommendations by United Nations Statistics Division. Since international trade in monetaryterms alone is not entirely representative of global trade from a transportation and logisticsperspective, the physical dimension of international trade is attaining increased importance.However, missing physical quantities in the trade flow databases, largely due to (1) noncompliance of reporter countries with the standard units of measurement or classification (2)confidentiality issues, or (3) erroneous collection and reporting of certain data, are observed.This paper, therefore, first presents the existing methods used in the literature to treat the issue ofmissing commodity weight information in international physical commodity trade databases, andthen proposes a stochastic multivariate imputation model, utilizing auxiliary variables such asmonetary trade data and price index, to impute missing physical quantities. Their relativeperformances in resolving the issue of incomplete physical commodity trade data are thenevaluated and compared through a case study, and it is concluded that the proposed approachoutperforms the existing approaches for commodity flow data imputation.