ResumoEste artigo aborda o tema da persistência na taxa de desemprego da região metropolitana de São Paulo. Foram utilizados modelos SARFIMA de integração fracionária para avaliar a dinâmica de absorção dos choques econômicos pela série de desemprego. A hipótese de histerese no desemprego foi avaliada usando os modelos de memória longa. Os resultados obtidos pelo arcabouço fracionário foram contrapostos aos do paradigma I (1) − I (0) e revelaram a incapacidade dos modelos tradicionais SARIMA de extrair corretamente o comportamento de baixa frequência da série temporal. Verificou-se que o modelo sazonal tradicional induz à superavaliação da persistência na série de tempo estudada. JEL classification: C22, C32, C15Palavras-chave: memória longa sazonal, integração fracionária e histerese. AbstractThis paper deals with the persistence theme in the unemployment rate of São Paulo metropolitan region. SARFIMA fractional integration models were used to evaluate the dynamics of economic shock absorption by the unemployment series. The hysteresis hypothesis in the unemployment rate was evaluated using long-memory models. The results found using the fractional framework were compared with those of the I (1)−I (0) paradigm, and they showed the inability of the traditional SARIMA models to correctly extract the low-frequency behavior of the time series. It was shown that traditional seasonal model leads to an overvaluation of the persistence in the series.
Este artigo tem por objetivo mostrar que as taxas de infl ação brasileira no período de jan(74 a jun(94 podem ter sido geradas por um processo com integração fracionária. Como essa categoria de processo estocástico não tem raiz unitária apesar de apresentar alta persistência, estaria assim explicado o comportamento da inflação considerado inconsistente por Cati, Garcia e Perron (1999). Considerando que a taxa de inflação pode ter sido gerada por um processo ARFIMA, estima-se sua ordem de integração por meio dos métodos da regressão do periodograma e da máxima verossimilhança. Em ambos os casos, os valores estimados para a ordem de integração são estatisticamente menores do que a unidade, indicando que o processo gerador da inflação brasileira tem integração fracionária, ou seja, apresenta longa persistência. AbstractThe ob jective of this article is to show that the Brazilian infl ation rates from January 1974 to June 1994 were generated by a fractionally integrated processo This process exhibits high persistency of shocks but does not have unit roots; these two characteristics can expIain the apparently inconsistent behavior ofthe infl ation pointed out by Cati, Garcia e Perron (1999). Assuming that the infl ation rates could have been generated by an ARFIMA model, we estimate its order of integration by maximum likelihood and the regression method. The results obtained in both cases.indicate that the inflation rates were generated by a long memory processo Palavras-Chave: Modelos ARFIMA, longa persistência, testes de raiz unitária, infl ação.
The purpose of this paper is to test the validity of the purchasing power parity (PPP) doctrine in Brazil. Historical data for the period 1855–1996 are considered. The period 1855–1990 is also analysed in order to compare the results with those obtained by Zini and Cati (1993) using the conventional cointegration analysis. This article uses fractional cointegration analysis, a flexible methodology which allows for more subtle forms of mean reversion. The tests performed are those of Geweke and Porter-Hudak (1983), and of Hurvich and Ray (1995). The critical values for both tests are generated by simulation because they are non-standard. The empirical results do not support the absolute PPP hypothesis but the relative PPP holds in the long run.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2025 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.