Research On The Improvement Of Data Audit Methods In International Comparison Project

As the largest international statistical cooperation project,the international comparison project(ICP)plays an important role in measuring the actual economic scale of various countries,comparing the price level of various countries and the well-being of residents.However,due to the influence of different economic development levels,consumption preferences,cultural customs,geographical environment and other factors,the specifications in the global ICP survey are difficult to meet the requirements of comparability,resulting in the results of international comparison divorced from the actual situation to a certain extent.In order to ensure the comparability of price data among regions,ICP began to increase the data review link in the 2011 round.However,for China,due to the late participation in ICP,the research on ICP,especially the data review content,is still in the primary stage;At the same time,due to China’s large geographical span and different regional economies in economy,culture and other aspects,the catalogue of ICP related specifications can not well match domestic consumption habits.Many reasons together lead to the difficulty of taking China’s actual situation into account in the data validation.Therefore,exploring and improving the existing problems of data validation tools can not only improve the reliability of international comparison results,but also have important practical significance in deepening China’s understanding of ICP and improving the voice in international comparison.Combined with the instructions on the classification of residents’ food expenditure in the ICP manual,this dissertation uses the price data of common agricultural products in the food category to study the validation ability of the ICP data validation tool Dikhanov table.This dissertation is divided into five chapters.The structure is organized as follows: the first chapter is the introduction,which combs the relevant literature,and advances layer by layer from the accuracy of ICP results,data quality and data validation research.The second chapter introduces the basic theory of data validation in detail.The third chapter introduces the theory and method of ICP data validation.It includes the basic information of data validation and the detailed combing of the content of data validation tools,in particular,the calculation method in data audit-CPD method is introduced,and further analyzes the possible defects of Dikhanov table in the process of practical application.The fourth chapter is the empirical part of the article,which makes an empirical test on the problems existing in the third chapter and the methods to solve these problems.The fifth part is the conclusion,suggestion and prospect.It summarizes and summarizes the results of the empirical research,evaluates the validation ability of Dikhanov table,points out the existing problems and puts forward corresponding policy suggestions.On this basis,it puts forward the prospect of the development of data validation.The starting point of this study is to find and improve the problems existing in the data validation methods.Therefore,the empirical research of this dissertation is divided into two parts: problem exploration and defect improvement.The empirical results of the first part are as follows: first,when the level of regional economic development is not considered,the validation will lead to too high results,and some real data will be excluded.Second,when the overall price change is artificially manipulated,the wrong price artificially adjusted is difficult to detect.Third,in the selection of representativeness of specification products,taking different experience as a reference,the selection of representativeness will lead to different change directions of validation results.At the same time,the validation results at the basic heading level are sensitive to the change of product representativeness.The second part of the demonstration is to propose solutions to the problems existing in the actual validation by adding economic adjustment factors,adding new validation standards and quantitatively selecting representative variables of specifications,and demonstrate the improved methods in combination with practice.The first is that the relevant economic factors can be adjusted to the real value to avoid errors in the empirical results.Second,the combination of PPP coefficient of variation and PG index is used to check the artificially adjusted wrong price,and the effect is obvious.Third,when determining the representativeness with reference to the actual price collection times,the validation results of different representative selection methods in vertical and horizontal have little difference,and have the same change trend,and the results are better than those of empirical selection.Based on the above conclusions,this dissertation puts forward relevant improvement suggestions that data validation should consider regional economic development differences,quantitative representative selection criteria and adding new validation indicators,in order to provide beneficial enlightenment for improving data validation ability and better ensure data quality.