ESTIMATION OF THE MISREPORTING MODELS USING MICRO-DATA SETS DERIVED FROM THE CONSUMER EXPENDITURE SURVEY: THE GAP BETWEEN MACRO AND MICRO ECONOMIC STATISTICS ON CONSUMER DURABLES

In many countries, a gap between macroeconomic and microeconomic statistics is observed. To explain the gap, the present paper tests the misreporting hypothesis originally proposed by Deaton and Irish [4]. The data used for estimation involves ten clusters of consumer durables from the Consumer Expenditure Survey in the US. Misreporting takes place, if a household purchased goods but did not report the amount (type 1 misreporting), or it purchased goods but reported the amount incorrectly (type 2 misreporting). The variance of the measurement error in type 2 misreporting is small and is not statistically