quinta-feira, setembro 30, 2010
terça-feira, setembro 28, 2010
Leitura do Dia - Credit models and the crisis, or: How I learned to stop worrying and love the CDOs
We follow a long path for Credit Derivatives and Collateralized Debt Obligations (CDOs) in particular, from the introduction of the Gaussian copula model and the related implied correlations to the introduction of arbitrage-free dynamic loss models capable of calibrating all the tranches for all the maturities at the same time. En passant, we also illustrate the implied copula, a method that can consistently account for CDOs with different attachment and detachment points but not for different maturities. The discussion is abundantly supported by market examples through history. The dangers and critics we present to the use of the Gaussian copula and of implied correlation had all been published by us, among others, in 2006, showing that the quantitative community was aware of the model limitations before the crisis. We also explain why the Gaussian copula model is still used in its base correlation formulation, although under some possible extensions such as random recovery. Overall we conclude that the modeling effort in this area of the derivatives market is unfinished, partly for the lack of an operationally attractive single-name consistent dynamic loss model, and partly because of the diminished investment in this research area.
Uma das raras discussões sérias sobre o assunto. O livro está em prioridade máxima para compra.
Meu comentário sobre o artigo da Wired sobre Cópulas Gaussianas em precificação de CDOs e a crise.
Artigos aprovados - 32 sbe
segunda-feira, setembro 27, 2010
Fifth Brazilian Conference on Statistical Modelling in Insurance and Finance
Destaque para os minicursos e convidados:
Nizar Touzi (Ecole Polytechnique, France)
Title: Stochastic Control Theory and Applications to Finance
Peter Tankov (Ecole Polytechnique, France)
Title: Pricing and Hedging Jump Risk in Financial Markets
Emiliano Valdez (University of Connecticut, USA)
José María Sarabia (University of Cantabria, Spain)
Narayanaswamy Balakrishnan (McMaster University, Canada)
Title: Some Recent Developments on Receiver Operating Characteristcs Analyses
Antonio Sanhueza (University of La Frontera, Chile)
Aristidis K. Nikoloulopoulos (University of East Anglia, UK)
Title: Vine Copulas with Asymmetric Tail Dependence and Applications to Financial Return Data
Dani Gamerman (Federal University of Rio de Janeiro, Brazil)
Title: Modeling Spatio-Temporal Dependence with Factor Models
Emily Fox (Duke University, USA)
Title: Dynamic Models for Covariance Matrices: A Rich New Class of Stationary Multivariate Volatility Process Models
Harry Joe (University of British Columbia, Canada)
Title: Vine Copula and Factor Copula Models for Financial Returns
Julio Stern (University of Sao Paulo, Brazil)
Title: The Rules of Logic Composition for Bayesian Epistemic e-Values
Victor Leiva (University of Valparaiso, Chile)
Title: On a Birnbaum-Saunders-t Fixed Effects Model and its Application to Insurance Data
sábado, setembro 25, 2010
O papel da imprensa livre.
Ator diz que "Sex and the City" não terá terceiro filme e culpa imprensa
Obrigado imprensa livre. Nos livrou da coisa mais estúpida desde o desenho dos smurfs.
quinta-feira, setembro 23, 2010
32 Encontro Brasileiro de Econometria - 2010
Generalized Empirical Likelihood/Minimum Contrast Estimation of Stochastic Differential Equations
Luiz Hotta; Márcio Laurini
Bayesian Inference Applied to Dynamic Nelson-Siegel Model with Stochastic Volatility in Factors
João Fróis Caldeira; Márcio Laurini; Marcelo Portugal
Estimation of stochastic volatility models using methods of generalized empirical likelihood/minimum contrast
Márcio Laurini; Luiz Hotta
Leitura do Dia - Education Performance: Was It All Determined 100 Years Ago? Evidence From São Paulo, Brazil
São Paulo, Brazil
Irineu de Carvalho Filho and Renato P. Colistete 1
This paper deals with institutional persistence in long-term economic development. We investigate the historical record of education in one of the fastest growing and most unequal societies in the twentieth century – the state of São Paulo, Brazil. Based on historical data from an agricultural census and education statistics, we assess the role played by factors such as land concentration, immigration and type of economic activity in determining supply and demand of education during the early twentieth century, and to what degree these factors help explain current educational performance and income levels. We find a positive and enduring effect of the
presence of foreign-born immigrants on the supply of public instruction, as well as a negative effect of land concentration. Immigrant farm-laborers established their own community schools, and pressured for public funding for those schools or for public schools. The effects of early adoption of public instruction can be detected more than one hundred years later in the form of better test scores and higher income per capita. These results are suggestive of an additional mechanism generating inequality across regions: the places that received immigration from countries with an established public education system benefited from an earlier adoption of the revolutionary idea of public education.
Irineu de Carvalho Filho is an Economist at the Research Department of the International Monetary Fund (e-mail: firstname.lastname@example.org); Renato P. Colistete is a Professor of Economic History at the Department of Economics of the Universidade de São Paulo (e-mail: email@example.com)
Muito, muito interessante. Além do tema fascinante e da qualidade da análise, a parte técnica é muito bem realizada. Artigo de destaque.
terça-feira, setembro 21, 2010
quinta-feira, setembro 16, 2010
Mini-curso de R - Estatística Unicamp
Dos dias 16 a 21 de setembro o departamento de Estatística da UNICAMP contará com a presença do professor Dr. John W. Emerson de Yale University, USA - Department of Statistics.
Para maiores informações sobre as áreas de pesquisa do professor. Veja
O programa das atividades é resumido em um Mini-curso sobre R com aplicações:
CONFERENCE TITLE: Topics in Data Analysis and Computational Statistics with R
SHORT CONFERENCE SUMMARY: This mini-conference will cover a wide range of topics at many different levels. First, a data analysis case study will provide an enjoyable exploration of nationalistic bias in Olympic judging, demonstrating the strengths of the R environment for work in applied statistics. The core sessions will introduce essential aspects of the R programming environment and advanced features of the package management system (including the C/C++ interface). The
concluding session will address scalable strategies for computing with massive data sets. Topics within the sessions will include Bayesian change point analysis; parallel programming with shared memory; and a few massive data examples (the Netflix data; the Airline On-time Arrival data; and efficient Landsat satellite data management with R).
FRIDAY 17 SEPTEMBER
14h00 Opening Data Analysis Case Study: "A Study of Olympic Judging."
Audience: anyone, wide audience, accessible to both students and researchers)
15h30 Coffee Break
16h00 Conference Overview: A preview of topics to be covered in the subsequent talks.
Audience: anyone who might come to any of the talks Monday/Tuesday.
MONDAY 20 SEPTEMBER
10h00 "Programming in R" -- an advanced introduction to the essentials of the R language.
Audience: anyone interested in R; not intended for advanced R users, though they are welcome.
14h00 "Bayesian Change Point Analysis and the R Package Management System."
Audience: some people with an interest in change point analysis, but also an important talk for R users and developers interested in the package management system and the C/C++ interface and parallel programming. Possible interest from bioinformatics and biostatistics and econometrics.
15h30 coffee Break
TUESDAY 21 SEPTEMBER
10h00 "Scalable Strategies for Computing with Massive Data Sets: The Bigmemory Project." Audience: R package developers and users with massive data sets or interested in topics relating to high-performance computing.
Pena que não vou poder ir, gostaria muito de assistir o seminário sobre o projeto Bigmemory.
quarta-feira, setembro 15, 2010
Notas na Capes
E parabéns também para o programa de Doutorado em Estatística da Unicamp, que em sua primeira avaliação estreou com 5 também. Pude acompanhar todo o esforço e dedicação dos professores para implementar o doutorado e foi uma experiência realmente marcante.
terça-feira, setembro 14, 2010
Artigo no ar
É o primeiro artigo a sair com a filiação do Ibmec RJ.
sexta-feira, setembro 10, 2010
O melhor é que essa segunda revisão permite contruir uma classe de estimadores bem mais gerais, e assim rende mais pesquisas.
terça-feira, setembro 07, 2010
Verdadeira revolução. Recentemente ligar na Cultura era assistir reprises intermináveis de programas antigos, ou pior, sobre o lixo das bandas de "rock" coloridas. Era deprimente.
Parabéns pela mudança.
segunda-feira, setembro 06, 2010
Leitura do dia - WARNING: PHYSICS ENVY MAY BE HAZARDOUS TO YOUR WEALTH!
to appear in Journal of Investment Management.
Andrew W. Lo and Mark T. Mueller
The quantitative aspirations of economists and financial analysts have for many years been based on the belief that it should be possible to build models of economic systems—and financial markets in particular—that are as predictive as those in physics. While this perspective has led to a number of important breakthroughs in economics, "physics envy" has also created a false sense of mathematical precision in some cases. We speculate on the origins of physics envy, and then describe an alternate perspective of economic behavior based on a new taxonomy of uncertainty. We illustrate the relevance of this taxonomy with two concrete examples: the classical harmonic oscillator with some new twists that make physics look more like economics, and a quantitative equity market-neutral strategy. We conclude by offering a new interpretation of tail events, proposing an "uncertainty checklist" with which our taxonomy can be implemented, and considering the role that quants played in the current financial crisis.
quinta-feira, setembro 02, 2010
Novo Working Paper - New Evidence on the Role of Cognitive Skill in Economic Development
Eduardo de Carvalho Andrade (Insper Institute of Education and Research)
Márcio Laurini (IBMEC Business School)
This paper presents new evidence on the role of cognitive skills in promoting economic growth rate. The novelties in this paper are that we use a within country dataset (Brazilian state and municipality levels data) and a new methodology, a nonparametric kernel regression estimation with mixed data. The main findings are the following: (i) there is strong evidence that the cognitive skill explains growth, but its relationship with growth appears to be non-linear, (ii) the quantity of schooling remains significant even after controlling for the quality of schooling, and (iii) there is support to the hypothesis that the effect of the cognitive skill on growth is important in an open environment.