Research of Methods for Construction of Supply and Use Tables Time Series

The supply, use and input-output tables are a powerful tool for economic analysis, but they are published only once in several years. An effective analysis of the economy often requires annual tables — tables time series. To construct such time series, special mathematical methods are used for tables projecting. The purpose of this study is to analyse the effectiveness of two such methods, the cycling method and the matrix transformation technique method. These methods have not yet been widely studied, so the results of this work can be used when choosing a method for constructing time series of supply, use and input-output tables. In the study, the selected projection methods were implemented in the Python programming language and were applied to tables of Russia and several other countries. After that, the projection results were compared and conclusions about the effectiveness of the methods were drawn. It turned out that the cycling method constructs the tables time series more accurately than the standard methods NRAS, INSD, Kuroda and the MTT method. In contrast, the MTT method was not very good at balancing time series. This paper contains 48 pages, 12 figures, 16 tables, 1 appendix. The list of sources used consists of 12 items. Keywords: supply, use, input-output tables, time series, cycling method, MTT method.