An explanation of the Dunning–Kruger effect is provided which does not require any psychological explanation, because it is derived as a statistical artifact. This is achieved by specifying a simple statistical model which explicitly takes the (random) boundary constraints into account. The model fits the data almost perfectly. JEL Classification: A22; C24; C91; D84; D91; I21

Confidence and overconfidence are essential aspects of human nature, but measuring (over)confidence is not easy. Our approach is to consider students' forecasts of their exam grades. Part of a student's grade expectation is based on the student's previous academic achievements; what remains can be interpreted as (over)confidence. Our results are based on a sample of about 500 second-year undergraduate students enrolled in a statistics course in Moscow. The course contains three exams and each student produces a forecast for each of the three exams. Our models allow us to estimate overconfidence quantitatively. Using these models we find that students' expectations are not rational and that most students are overconfident, in agreement with the general literature. Less obvious is that overconfidence helps: given the same academic achievement students with larger confidence obtain higher exam grades. Female students are less overconfident than male students, their forecasts are more rational, and they are also faster learners in the sense that they adjust their expectations more rapidly.

Neste trabalho analisou-se a estrutura dos sistemas meteorológicos associados às precipitações intensas em Alagoas entre 2003 e 2006. Foi feita a análise das imagens de satélite no canal infravermelho, de produtos de reanálise do National Centers for Environmental Prediction (NCEP) e dados de precipitação de dezessete postos pluviométricos, espalhados em diferentes regiões do Estado. O objetivo desse estudo foi analisar as causas da formação das precipitações intensas e determinar os sistemas meteorológicos associados. Os casos de precipitação intensa foram registrados no Outono/Inverno. Além disso, metade dos eventos ocorridos entre a primavera e o verão aconteceu em janeiro. As Perturbações Ondulatórias nos Ventos Alísios, a Corrente de Jato do Nordeste Brasileiro e as extremidades das frentes frias foram os sistemas que estiveram mais relacionados às precipitações intensas. Complexos Convectivos de Mesoescala provocaram boa parte dos temporais no oeste do Estado, o que não foi verificado no leste. Exemplificando o método de análise utilizado, um caso foi discutido.