The number of Covid-19 cases is increasing dramatically worldwide, with several countries experiencing a second and worse wave. Therefore, the availability of reliable forecasts for the number of cases and deaths in the coming days is of fundamental importance. We propose a simple statistical method for short-term real-time forecasting of the number of Covid-19 cases and fatalities in countries that are latecomers  i.e., countries where cases of the disease started to appear some time after others. In particular, we propose a penalized (LASSO) regression with an error correction mechanism to construct a model of a latecomer in terms of the other countries that were at a similar stage of the pandemic some days before. By tracking the number of cases in those countries, we forecast through an adaptive rolling-window scheme the number of cases and deaths in the latecomer. We apply this methodology to four dierent countries: Brazil, Chile, Mexico, and Portugal. We show that the methodology performs very well. These forecasts aim to foster a better short-run management of the health system capacity and can be applied not only to countries but to dierent regions within a country, as well.

We study the asymmetric effects of procyclical and countercyclical expansions of public banks’ credit on economic growth. Using a panel of Brazilian municipalities (2009-2014) and the same identification strategy as Greenstone et al. (2020), we show that the effect of public credit on the economic performance of Brazilian municipalities was more substantial in 2009, the only year in our sample in which the public credit policy was countercyclical. Interestingly, our estimates also suggest that the effect of credit policies on growth is more salient when the banking market is more concentrated. Guided by the empirical results we built a theoretical model, which is calibrated with Brazilian data. Policy experiments based on the model demonstrate that countercyclical public credit policies are more effective than procyclical public credit policies. In line with the empirical results, the theoretical model also shows that public credit policies in general (either in booms or in recessions) are more efficient when credit markets are more concentrated. This result indicates that the structure of credit markets is crucial to explain the impact of public credit policies on growth.

In this paper we survey the most recent advances in supervised machine learning and highdimensional models for time series forecasting. We consider both linear and nonlinear alternatives. Among the linear methods we pay special attention to penalized regressions and ensemble of models. The nonlinear methods considered in the paper include shallow and deep neural networks, in their feed-forward and recurrent versions, and tree-based  methods, such as random forests and boosted trees. We also consider ensemble and hybrid models by combining ingredients from different alternatives. Tests for superior predictive ability are briefly  reviewed. Finally, we discuss application of machine learning in economics and finance and provide an illustration with high-frequency financial data

A paridade do poder de compra da moeda (“PPP”), uma das teorias mais discutidas no meio acadêmico, sustenta que a taxa de câmbio nominal entre duas moedas deve ser igual à relação dos níveis de preços agregados entre os dois países, de modo que uma unidade de moeda de um país terá o mesmo poder de compra em um país estrangeiro. O objetivo deste trabalho é investigar a validade da teoria de PPP e a inovação que trazemos para a literatura é a aplicação desta teoria para o desenvolvimento de uma estratégia quantitativa de compra/venda de pares de moedas utilizando as séries diárias de preços calculadas pela PriceStats. Os resultados encontrados aqui sugerem que, apesar de identificarmos uma relação entre o diferencial de inflação e movimentos da taxa de câmbio nominal nas séries do PriceStats, observamos um desempenho quantitativo pior das nossas estratégias de investimento em moedas baseadas no modelo de Paridade do Poder de Compra da Moeda (PPP) vis à vis outros modelos padrão dentro da literatura financeira. Por outro lado, para pares específicos de moedas, encontramos números interessantes quando baseamos nossa estratégia nos modelos de PPP, observando Hit Ratio superior a 50% e retorno acumulado positivo da estratégia.

This dissertation studies the effects of the fiscal stance on the composition of public debt in the short run. We use data on Brazilian public debt issuance and assess the impact of fiscal deficits and sovereign risk on the share of shortterm debt through reduced-form and VAR methods. Our results suggest that a fiscal deterioration is associated with a higher share of short-term debt. At the same time, a sovereign risk shock increases the reliance on short-term and floating-rate debt. Then, in order to disentangle supply and demand factors in public debt issuance, we estimate the interest-rate elasticity in auctions for short- and medium-term public debt. Using a method of identification through heteroskedasticity, we find that both factors are present. However, market demand is considerably more interest-rate elastic than Treasury supply.

A evidência empírica demonstra que termos de troca e produtividade flutuam juntos para economias grandes exportadoras de commodities. O modelo semi estrutural estimado nesta dissertação visa representar uma pequena economia aberta, na qual as flutuações dos termos de troca impactam de forma direta a produtividade da economia. A partir de métodos Bayesianos, o modelo testa se existe correlação positiva entre períodos de expansão dos termos de troca e períodos de expansão da produtividade, com o objetivo de analisar a influência dos termos de troca na dinâmica da economia brasileira de 2000 a 2019. Os resultados obtidos corroboram a correlação positiva prevista nos dados.

We develop a two-period overlapping generations model in which both the family structure and the decision to commit crime are endogenous and the dynamics of moral norms of good conduct is transmitted intergenerationally by families and peers. By "destroying" biparental families and putting fathers in prison, we show that more intense crime repression can backfire because it increases the possibility that criminals' sons become criminals themselves. Our model also explains the emergence and persistence of urban ghettos characterized by a large proportion of broken families, high crime rates, and high levels of peer socialization, which reinforce criminal activities

The existing literature on the effects of Covid on stock returns focuses on endogenous changes in risk tolerance and on the modeling of rare events. So far, these attempts have not been able to match the data. In this paper, I propose an alternative approach to explaining the Covid effects on stock returns worldwide: disentangling the long-run effects from the short-run effects. Intuitively, Covid’s long-run effects include disruptions of supply chains and educational patterns, which, conceivably, will take time to phase out. Exactly as it happens with the persistent shocks of long-run risks models! A model that allows for short-run fluctuations and long-run risk shows that persistent shocks play a role in explaining stock market returns and exchange rates in a time span that starts in January 2018 and ends in November 2021.

Estagnação econômica, dependência exclusiva da exportação de matérias primas agrícolas, uma vasta plantation escravista, governos irremediavelmente deficitários. Estes são alguns dos fatos estilizados que vêm à mente quando se pensa na história econômica do Brasil Império. Com base em ampla evidência estatística e em diálogo permanente com a historiografia clássica e a produção acadêmica contemporânea, este livro confirma que tal espécie de “fatos” costuma ser apenas parcialmente verdadeira. A economia do Império não era imóvel – movia-se, ainda que a passos lentos

This thesis is composed of three papers on financial econometrics. The first two papers explore the relation between intraday equity market returns and implied volatility, represented by the CBOE Volatility Index (VIX). In both papers, we estimate one-minute-ahead forecasts using rolling windows within a day. In the first paper, the estimates indicate that our volatility factor models outperform traditional benchmarks at high frequency time-series analysis, even when excluding crisis periods. We also find that the model has a better out-of-sample performance at days without macroeconomic announcements. Interestingly, these results are amplified when we remove the crisis period. The second paper proposes a machine learning modeling approach to this forecasting exercise. We implement a minute-by-minute rolling window intraday estimation method using two nonlinear models: Long-Short-Term Memory (LSTM) neural networks and Random Forests (RF). Our estimations show that the VIX is the strongest candidate predictor for intraday market returns in our analysis, especially when implemented through the LSTM model. This model also improves significantly the performance of the lagged market return as predictive variable. Finally, the third paper explores a multivariate extension of the FarmPredict method, by combining factor-augmented vector autoregressive (FAVAR) and sparse models in a high-dimensional environment. Using a three-stage procedure, we estimate and forecast factors and its  oadings, which can be observed, unobserved, or both, as well as a weakly sparse idiosyncratic structure. We provide an application of this methodology to a panel of daily realized volatilities. Finally, the accuracy of the stepwise method indicates improvements of this forecasting method when compared to consolidated benchmarks.

Nowcasting in economics is the prediction of the present, the recent past or even the prediction of the very near future of a certain indicator. Generally, a nowcast model is useful when the value of a target variable is released with a significant delay with respect to its reference period and/or when its value gets notably revised over time and stabilizes only after a while. In this thesis, we develop and analyze several Nowcasting methods using high-dimensional (big) data in different contexts: from the forecasting of economic series to the nowcast of COVID-19. In one of our studies, we compare the performance of different Machine Learning algorithms with more naive models in predicting many economic variables in real-time and we show that, most of the time, Machine Learning beats benchmark models. Then, in the rest of our exercises, we combine several nowcasting techniques with a big dataset (including high-frequency variables, such as Google Trends) in order to track the pandemic in Brazil, showing that we were able to nowcast the true numbers of deaths and cases way before they got available to everyone.

As economies develop and grow, their informal sector shrinks. The literature emphasizes a number of supply-side causes (higher costs of informality for larger and capital-intensive firms, improved state enforcement capacity, higher levels of education) to explain this phenomenon. This thesis contributes to the debate by proposing a new, demand-side explanation. We argue that the rise in formality can be explained, in part, by a rise in demand for formal goods and services from households whose income is growing. Using Brazilian household expenditure survey data, we document that in the cross-section, higher-earning households consume a larger fraction of formal goods (7 percentage points as income doubles). We also show that, over time, formal consumption increases together with income. We attempt to provide a causal estimate by analysing exogenous increases in the minimum wage. Last, we propose a theoretical discussion on the type of preferences consistent with this observed behavior.