Simple monetary policy rules and data mining methods

The paper deals with the problem of modelling simple monetary policy rules, example of which is the Taylor rule, which is widely present in the literature of macroeconomics. More precisely, the aim of this paper is realized through verification whether the monetary policy that is carried out by central banks (reduced to setting of nominal short-term interest rate) may be in a statistically significant way modelled in the form of simple decision rules based on macroeconomic variables. Apart from generating monetary policy rules, the paper also aims at verifying the impact of global economic crisis of 2008-2009 on their stability as well as identifying the differences between monetary policy rules for various central banks. The paper makes use of heuristic methods of data analysis (data mining) as they enable identification of relations in data sets without preliminary hypotheses, which is the aim of this paper. The cases of three central banks are analyzed- the National Bank of Poland, the Bank of England and the US Federal Reserve. The best rules as regards both statistical significance and economic interpretation are rules generated for the Federal Reserve, whereas the results for the other central banks are unsatisfactory. It was also established that simple monetary policy rules are significantly destabilized in the period of the global economic crisis of 2008-2009.