Methods for estimating loss of relevant information of the time rows relative to their length

The analyzing process of economic time series is accompanied by the problem of small samples. In the present work the adequacy of information in a predetermined volume of sample is estimated. Relevance of the information is evaluated using statistical and fractal analysis tools. The authors also used mathematical apparatus of the theory of fuzzy sets. In contrast to the known results of the statistical analysis fractal analysis provides an estimate of such new features of time series as presence of memory and estimation of its depth. Sufficiency of lengths of investigated time rows is estimated by depth of their memory.