O regime de ciclos de aprendizagem e a heterogeneidade de seus efeitos sobre a proficiência dos alunos

Neste artigo fazemos a avaliação do impacto da adoção do regime de ciclos ao longo da distribuição intra-classe de proficiência em matemática e em língua portuguesa. A população de interesse é o das crianças matriculadas nas 4ª e 8ª séries (5º e 9º ano). Os resultados indicam que há fraca relação entre a mudança de um regime seriado para o de ciclos sobre o desempenho, seja em média ou ao longo da distribuição. Contudo, entre escolas que adotaram o sistema de ciclos no passado, mas que retornaram ao regime seriado, a adoção das séries gerou impactos positivos no desempenho de matemática na 4ª série para todos os pontos da distribuição de desempenho. Pode-se concluir que, em geral, a reprovação não parece ser um mecanismo forte para gerar aumentos de esforço dos alunos, sobretudo entre alunos mais velhos.

Analysis of contagion from the constant conditional correlation model with Markov regime switching

Over the last decades, the analysis of the transmissions of international nancial events has become the subject of many academic studies focused on multivariate volatility models volatility. The goal of this study is to evaluate the nancial contagion between stock market returns. The econometric approach employed was originally presented by Pelletier (2006), named Regime Switching Dynamic Correlation (RSDC). This methodology involves the combination of Constant Conditional Correlation Model (CCC) proposed by Bollerslev (1990) with Markov Regime Switching Model suggested by Hamilton and Susmel (1994). A modi cation was made in the original RSDC model, the introduction of the GJR-GARCH model formulated in Glosten, Jagannathan e Runkle (1993), on the equation of the conditional univariate variances to allow asymmetric e ects in volatility be captured. The database was built with the series of daily closing stock market indices in the United States (SP500), United Kingdom (FTSE100), Brazil (IBOVESPA) and South Korea (KOSPI) for the period from 02/01/2003 to 09/20/2012. Throughout the work the methodology was compared with others most widespread in the literature, and the model RSDC with two regimes was de ned as the most appropriate for the selected sample. The set of results provide evidence for the existence of nancial contagion between markets of the four countries considering the de nition of nancial contagion from the World Bank called very restrictive. Such a conclusion should be evaluated carefully considering the wide diversity of de nitions of contagion in the literature.