Abstract

The objective of this work was to fit a distribution to the IPCA12 dataset and estimate the probability of this index remaining within the confidence limits established by the Central Bank of Brazil for year 2024, which are 3% +- 1.5%. To choose between the log-normal, Gamma, and Weibull distributions, Kolmogorov-Smirnov test, and the Akaike Information Criterion (AIC) were analyzed. The Gamma model with parameters alfa 5.81 and beta 1.09 was selected, and it was estimated that the probability of the true value of IPCA staying within the confidence interval established by the Central Bank for the year 2024 would be 25.45%. Furthermore, maintaining a margin of +- 1.5%, it was possible to conclude that the IPCA value or target that would maximize coverage of the range should be 5.4% instead of 3%. More specifically: P(5.4% - 1.5% <= IPCA12 <= 5.4% + 1.5%) = 45.94%.

Keywords

• Asymmetry
• kurtosis
• maximum likelihood

References

• AKAIKE, H. A new look at the statistical model identification. IEEE Transactions on
• Automatic Control, Notre Dame, v. 19, n. 6, p. 717-723, 1974.
• BARROS, R. P. de; NERI, M. M., R. Pobreza e inflação no Brasil: Uma análise agregada.
• Economia Brasileira em Perspectiva. Rio de Janeiro: IPEA, vol 1, 1996. 838 p.
• CONOVER, W. J. Practical Nonparametric Statistical. John Wiley & Sons Inc., New York,
• p.
• DELFIM NETTO, A. Sobre as metas inflacionárias. Revista de Economia Aplicada, v. 3, n. 3,
• EMILIANO, P. C.; VIVANCO, M. J. F.; MENEZES, F. S. Information criteria: How do they
• behave in different models?. Computational Statistics & Data Analysis, v. 69, p. 141-153, 2014.
• FERREIRA, T. P.; FIGUEIREDO F. M. R. Os preços administrados e a inflação no Brasil.
• Banco Central do Brasil Trabalhos para Discussão, n. 59, Brasília, dez. 2002.
• FERREIRA, T.P.; PETRASSI,M. B. S. Regime de metas de inflação: resenha sobre a
• experiência internacional. Banco Central do Brasil. Notas Técnicas, no 30, Brasília, nov., 2002.
• MARSAGLIA G., TSANG W. W., WANG J. Evaluating Kolmogorov’s distribution. Journal of
• Statistical Software, v. 8, n. 18, 2003.
• MILLER, R. B. Maximum likelihood estimation and inference: with examples in R, SAS, and
• ADMB. 1. ed. [S.l.]: Wiley, 2011.
• MOOD, A. M.; GRAYBILL, F. A.; BOES, D. C. Introduction to the theory of statistics.
• Singapore: McGraw-Hill, 1974. 564 p.
• R CORE TEAM. R: a language and environment for statistical computing. R Foundation for
• Statistical Computing, Vienna, Austria. 2023.

Paper information