April 12, 2007

Modelling the reporting discrepancies in bilateral data

This paper is about the discrepancies in reported bilateral statistical data ("mirror data"). For example the trade from country A to country B is not reported the same in the two countries.

The discrepancies are used to estimate the accuracy of the reporters. The estimated accuracies are to be used to compute optimal combinations of mirror data.

Two models of the discrepancies are presented: (a) unbiased reporting with inaccurate reporters having a large variance, and (b) biased reporting with inaccurate reporters having a large bias (either positive or negative). Estimation methods are least squares regression and maximum likelihood.

A numerical illustration is given, using data of the international trade in services. It is shown how to judge the two models empirically.

For an updated version, see CPB Discussion Paper 216.