1 Quality measures in non-statistical sampling: MFI Interest Rates Statistics (MIR)

Traditional literature on sampling techniques mainly focuses on statistical samples and only marginally covers non-random (non-statistical) samples. Nevertheless, there has been a recent revival in interest on non-statistical samples due to their widespread use in certain fields like government surveys, marketing research, or for audit purposes. This paper tries to set up common rules for non-statistical sample in which only those largest institutions within each stratum are collected. This is done by focusing on the regular collection of Monetary Financial Institutions (MFIs) interest rate statistics from countries of the European Union as they are collected by the European System of Central Banks (ESCB). The paper concludes by proposing a way to establish common rules for non-statistical samples based on the basis of a synthetic measurement of a mean of absolute errors