Modeling the Relationship Between Timeliness and Quality of Primary Statistical Data Using Bernstein Pair-Copula Constructions

The article presents the results of a study on the relationship between the timeliness and accuracy of primary statistical data. The problem of finding a balance between the deadlines for collecting reports and the quality of information provided by organizations is considered. The aim of the paper is to quantitatively test the assumption that an increase in the time available for completing reports improves their quality; the relevance is due to the need to justify the deadlines for timely statistical reporting. The study uses data from form No. P-1 «Information on the Production and Shipment of Goods and Services» for enterprises in Moscow and the Moscow Region for 2023; the sample comprised 1725 observations; resampling was applied to test whether the results could be generalized to the target population. The feasibility of generalizing the results to the population has been demonstrated, including through the use of resampling. The author proposes constructing a multivariate distribution of indicators using pair-copula constructions and Bernstein copulas, while estimating marginal distributions using empirical distribution functions. A statistically significant but weak positive correlation was found be tween the time available for completing reports and the quality of primary data ( τ ≈ 0.07), which does not confirm the initial assumption that quality improves with longer reporting deadlines. The practical significance is related to the possibility of taking the findings into account when setting reporting deadlines. The robustness of the model is demonstrated; the small magnitude of the correlation calls for cautious management decisions and further study of factors affecting the quality of primary statistical information.