Methodological problems of economics as a nominalistic science

This article focuses on a description of a substantial methodological change in economics that has been being in progress for a previous couple of decades. To explain this change the article exploits the bases of K. R. Popper's distinction between nominalism and essentialism in science. The article comes to a conclusion that the methodological essence of economics in recent years has been shifting from the essentialist shape to the nominalistic one. The passage from essentialism to nominalism is closely connected with the development and the extension of quantitative methods, statistics and advanced mathematics in economics. Another characteristic of this passage leans on the increasing ability of economics to invent and offer hypotheses testable by Popper's falsification method. Main purpose of the article is to describe and analyse some of the problems connected with mentioned methodological turn in economics. That is why problems like data-mining or biased tests are deeply analysed in this paper and methods of avoiding this problems are proposed. The last part of the paper comes to the conclusion that thanks to described methodological turn in economics this science is neither only inductive-based nor just deductive-based but the methodology of contemporary economics is rather the compromise of both. This compromise creates a new methodological paradigm for economic research.