“Permutation entropy and information recovery in nonlinear dynamic economic time series” is an article written by Miguel Henry and George Judge in 2019, and published in the March issue of Econometrics journal. The article presents an insight into a logical approach to the process of information extraction from complex economic systems. Permutation entropy (PE) is a method that is used to define the pattern-ordinal relations of time series, which could be characterized by the high speed of computation and simplicity of its concept.
The authors address the problem of using PE to explore fundamental themes, such as complexity and predictability, powerfully and effectively. To cover the issue, the authors defined the methodology of using PE and developed a framework for using PE in the Dow Jones Industrial Average time series system (Henry and Judge 3). The paper features a section dedicated to the empirical results achieved through the application of the methodology on the data from periods post World War II. The article’s conclusions show that PE could be utilized to extract qualitative information from a wide range of economic time series, including regular, noisy, and even chaotic types.
The article presents an innovative approach to the subject of PE use for data extraction. Moreover, the article addresses the possible applications of the methodology, supported by the evidence base. I find it helpful that the authors featured the potential applications for the methodology as it illustrates the method’s versatility and effectiveness. As a critique, even though I do not think that the authors made any mistakes or provided any confusing information, the article could feature a comparison with other alternative measures. The authors also needed to address the differences between micro and macroeconomic systems in the use of conducted PE framework. As a minor comment, while there are no grammatical errors in the paper, the quote from a decade ago as the primary point of introduction could be replaced by more relevant information.
Work Cited
Henry, Miguel, and George Judge. “Permutation Entropy and Information Recovery in Nonlinear Dynamic Economic Time Series.” Econometrics, vol. 7, no. 1, pp. 1-16.