Year: 2025
Type: article
Abstract: Purpose Existing classifications of inflationary regimes often rely on subjective judgments, hindering objectivity and accuracy. This study proposes a novel, data-driven approach to address this limitation. Design/methodology/approach We combine unsupervised clustering and classification tree methods to analyze Argentine inflation data from 1943 to 2022. Two smoothing techniques are introduced: a measure of temporal contiguity and a rolling majority rule method. The resulting regimes are compared to existing classifications based on their explanatory power for inflation-relative price variability. Findings Our method identifies distinct inflationary regimes, demonstrating significant improvement in objectivity and accuracy compared to existing literature. The regimes capture key historical periods and exhibit a strong association with inflation-relative price variability, providing valuable insights into Argentine inflation dynamics. Originality/value This study offers a novel methodological framework for constructing objective and accurate inflationary regimes, free from subjective biases. This approach holds potential for application to other contexts and contributes to a more nuanced understanding of inflation dynamics.
Source: EconomiA
Authors: Manuel de Mier, Fernando Delbianco
Language: English
Domain: Social Sciences
Open Access status: diamond
OpenAlex ID: W4407341291