Forecasting electricity load demand: analysis of the 2001 rationing period in Brazil

This paper studies the electricity load demand behavior during the 2001 rationing period, which was implemented because of the Brazilian energetic crisis. The hourly data refers to a utility situated in the southeast of the country. We use the model proposed by Soares and Souza (2003), making use of generalized long memory to model the seasonal behavior of the load. The rationing period is shown to have imposed a structural break in the series, decreasing the load at about 20%. Even so, the forecast accuracy is decreased only marginally, and the forecasts rapidly readapt to the new situation. The forecast errors from this model also permit verifying the public response to pieces of information released regarding the crisis.

Estimating and forecasting the volatility of Brazilian finance series using arch models (Preliminary Version)

The goal of this paper is to present a comprehensive emprical analysis of the return and conditional variance of four Brazilian …nancial series using models of the ARCH class. Selected models are then compared regarding forecasting accuracy and goodness-of-…t statistics. To help understanding the empirical results, a self-contained theoretical discussion of ARCH models is also presented in such a way that it is useful for the applied researcher. Empirical results show that although all series share ARCH and are leptokurtic relative to the Normal, the return on the US$ has clearly regime switching and no asymmetry for the variance, the return on COCOA has no asymmetry, while the returns on the CBOND and TELEBRAS have clear signs of asymmetry favoring the leverage e¤ect. Regarding forecasting, the best model overall was the EGARCH(1; 1) in its Gaussian version. Regarding goodness-of-…t statistics, the SWARCH model did well, followed closely by the Student-t GARCH(1; 1)

Model selection, estimation and forecasting in VAR models with short-run and long-run restrictions

We study the joint determination of the lag length, the dimension of the cointegrating space and the rank of the matrix of short-run parameters of a vector autoregressive (VAR) model using model selection criteria. We consider model selection criteria which have data-dependent penalties as well as the traditional ones. We suggest a new two-step model selection procedure which is a hybrid of traditional criteria and criteria with data-dependant penalties and we prove its consistency. Our Monte Carlo simulations measure the improvements in forecasting accuracy that can arise from the joint determination of lag-length and rank using our proposed procedure, relative to an unrestricted VAR or a cointegrated VAR estimated by the commonly used procedure of selecting the lag-length only and then testing for cointegration. Two empirical applications forecasting Brazilian in ation and U.S. macroeconomic aggregates growth rates respectively show the usefulness of the model-selection strategy proposed here. The gains in di¤erent measures of forecasting accuracy are substantial, especially for short horizons.

On the nature of income inequality across nations

In this paper, we investigate the nature of income inequality across nations. First, rather than functional forms or parameter values in calibration exercises that can potentially drives results, we estimate, test, and distinguish between types of aggregate production functions currently used in the growth literature. Next, given our panel-regression estimates, we perform several exercises, such as variance decompositions, simulations and counter-factual analyses. The picture that emerges is one where countries grew in the past for different reasons, which should be an important ingredient in policy design. Although there is not a single-factor explanation for the difference in output per-worker across nations, inequality, followed by distortions to capital accumulations and them by human capital accumulation.

Model selection, estimation and forecasting in VAR models with short-run and long-run restrictions

We study the joint determination of the lag length, the dimension of the cointegrating space and the rank of the matrix of short-run parameters of a vector autoregressive (VAR) model using model selection criteria. We suggest a new two-step model selection procedure which is a hybrid of traditional criteria and criteria with data-dependant penalties and we prove its consistency. A Monte Carlo study explores the finite sample performance of this procedure and evaluates the forecasting accuracy of models selected by this procedure. Two empirical applications confirm the usefulness of the model selection procedure proposed here for forecasting.

Sacrifice and efficiency of the income tax schedule

We investigate the efficiency of equal sacrifice tax schedules in an economy which primitives are exactly those in Mirrlees (1971): a continuum of individuals with identical preferences defined over consumption and leisure who differ with respect to their labor market productivity. Using a separable specification for preferences we derive the minimum equal sacrifice allocation and recover the tax schedule that implements it. The separable specification allows us to use the methodology developed by Werning (2007b) to check whether the schedule is efficient, that is, whether there is no alternative tax schedule that raises more revenue while delivering less utility to no one. We find that inefficiency does not arise for most parametrizations we use to approximate the US economy. For the few cases for which inefficiency does arise, it does so only for very high levels of income and marginal tax rates.

On the efficiency of equal sacrifice income tax schedules

In an economy which primitives are exactly those in Mirrlees (1971), we investigate the efficiency of labor income tax schedules derived under the equal sacrifice principle. Starting from a given government revenue level, we use Werning’s (2007b) approach to assess whether there is an alternative tax schedule to the one derived under the equal sacrifice principle that raises more revenue while delivering less utility to no one. For our preferred parametrizations of the problem we find that inefficiency only arises at very high levels of income. We also show how the multipliers of the Pareto problem may be extracted from the data and used to find the implicit marginal social weights associated with each level of income.

Forecasting multivariate time series under present-value-model short- and long-run co-movement restrictions

It is well known that cointegration between the level of two variables (e.g. prices and dividends) is a necessary condition to assess the empirical validity of a present-value model (PVM) linking them. The work on cointegration,namelyon long-run co-movements, has been so prevalent that it is often over-looked that another necessary condition for the PVM to hold is that the forecast error entailed by the model is orthogonal to the past. This amounts to investigate whether short-run co-movememts steming from common cyclical feature restrictions are also present in such a system. In this paper we test for the presence of such co-movement on long- and short-term interest rates and on price and dividend for the U.S. economy. We focuss on the potential improvement in forecasting accuracies when imposing those two types of restrictions coming from economic theory.

The performance of open-end Brazilian fixed income mutual funds for retail clients

From a financial perspective, this dissertation analyzes the Brazilian mutual fund industry performance for an average retail client. The most representative funds for the local population, that are the fixed income open-end ones, will be selected and their performance will be measured aiming to answer if clients of this industry obtained a proper return over their investments in the period between August 2010 and August 2013. A proper return will be understood as the preservation of the purchasing power of the individual´s savings, what is achieved with a positive performance of a mutual fund after discounting taxes, administrative fees and inflation. After obtaining an answer for the previous question, this dissertation will explore a possible alternative solution: Tesouro Direto, that is an example of a financial approach that could foster the disintermediation between savings and investments through electronic channels. New electronic platforms, with a broader scope, could be utilized to increase the efficiency of funding productive investments through better remunerating Brazilian savings. Tesouro Direto may point towards a new paradigm.

Forecasting Brazilian inflation by its aggregate and disaggregated data: a test of predictive power by forecast horizon

This work aims to compare the forecast efficiency of different types of methodologies applied to Brazilian Consumer inflation (IPCA). We will compare forecasting models using disaggregated and aggregated data over twelve months ahead. The disaggregated models were estimated by SARIMA and will have different levels of disaggregation. Aggregated models will be estimated by time series techniques such as SARIMA, state-space structural models and Markov-switching. The forecasting accuracy comparison will be made by the selection model procedure known as Model Confidence Set and by Diebold-Mariano procedure. We were able to find evidence of forecast accuracy gains in models using more disaggregated data

Estimation of random coefficients logit demand models: an application to the Brazilian fixed income fund market

Estimation of demand and supply in differentiated products markets is a central issue in Empirical Industrial Organization and has been used to study the effects of taxes, merges, introduction of new goods, market power, among others. Logit and Random Coefficients Logit are examples of demand models used to study these effects. For the supply side it is generally supposed a Nash equilibrium in prices. This work presents a detailed discussion of these models of demand and supply as well as the procedure for estimation. Lastly, is made an application to the Brazilian fixed income fund market.

Income inequality and human capital development

Investimentos em capital humano são essenciais para o desenvolvimento econômico de um pais. No Brasil, diversas fontes apontam para a falta de mão de obra qualificada como sendo uma das causas de um fraco crescimento econômico. Esta dissertação explora as teorias que ligam desigualdade de renda com performance econômica. A parte empírica se foca em uma das teorias apresentadas, a de imperfeições no mercado de credito. De acordo com esta teoria, mercados de credito imperfeitos são fracos alocadores de recursos e não possibilitam que indivíduos de baixa renda invistam no próprio capital humano. No Brasil, há uma escassez de estudos empíricos focados em testar os canais através dos quais a desigualdade de renda afeta o crescimento, trazendo significância para esta dissertação. Os resultados apresentados aqui foram obtidos através da pesquisa familiar – POF – realizada pelo IBGE. Os dados mostram que investimentos em educação crescem como percentual do orçamento com o aumento da renda familiar. Aumentos de renda para classes de renda já elevadas não provocam igual aumento nas despesas educacionais. Os dados sugerem a existência de uma restrição orçamentária para Brasileiros de baixa e média renda independente da região. Foram encontradas fortes evidencias de que classes de baixa e média renda no Brasil tem acesso limitado ao mercado de credito. Portanto, existe evidencia de que redistribuição aumentaria o gasto agregado em educação

Credit borrowing constraints in a DSGE framework: income vs. housing

Credit market in Brazil distinguishes from advanced economies in many aspects. One of them is related to collaterals for households borrowing. This work proposes a DSGE framework, based on Gerali et al.(2010), to analyse one pecularity of Brazillian credit market: payroll-deducted personal loans. To original model, we added the possibility to households contract long term debt and compare to differents types of credit constrains: one based on housing and other based on future income. We callibrate and estimate the model to Brazil, using Bayesian technique. Results show that, in a economy where credit constraints are based on income, responses to shocks appear to be stronger, at first, but dissipate faster. This occurs because income responds quickly to shock than housing prices, so does amount available to loans. In order to smooth consumption, agents compensate lower income and borrowing by increasing working hours, restoring loans and debt in a shorter time.

Forecasting Multivariate Time Series under Present-Value-Model Short- and Long-run Co-movement Restrictions

This paper has two original contributions. First, we show that the present value model (PVM hereafter), which has a wide application in macroeconomics and fi nance, entails common cyclical feature restrictions in the dynamics of the vector error-correction representation (Vahid and Engle, 1993); something that has been already investigated in that VECM context by Johansen and Swensen (1999, 2011) but has not been discussed before with this new emphasis. We also provide the present value reduced rank constraints to be tested within the log-linear model. Our second contribution relates to forecasting time series that are subject to those long and short-run reduced rank restrictions. The reason why appropriate common cyclical feature restrictions might improve forecasting is because it finds natural exclusion restrictions preventing the estimation of useless parameters, which would otherwise contribute to the increase of forecast variance with no expected reduction in bias. We applied the techniques discussed in this paper to data known to be subject to present value restrictions, i.e. the online series maintained and up-dated by Shiller. We focus on three different data sets. The fi rst includes the levels of interest rates with long and short maturities, the second includes the level of real price and dividend for the S&P composite index, and the third includes the logarithmic transformation of prices and dividends. Our exhaustive investigation of several different multivariate models reveals that better forecasts can be achieved when restrictions are applied to them. Moreover, imposing short-run restrictions produce forecast winners 70% of the time for target variables of PVMs and 63.33% of the time when all variables in the system are considered.

Social security effects on income distribution: a counterfactual analysis for Brazil

According to Diamond (1977), one of the reasons for the existence of social security systems is that they function as an income redistribution mechanism. There is an extensive literature that tests whether social security systems produce the desired results in developed countries (mainly for the U.S.A.). Nevertheless, there is not an obvious consensus about this social security property and there is little evidence for developing countries. In this article, we test this property for the Brazilian Social Security System. In addition, we also look at another question which has not been answered yet in the previous literature. Is the trend of social security systems increasingly progressive or regressive? We conclude that the changes in Brazilian Social Security legislation reduced inequality between 1987 and 1996, but only for the elderly. For the other age groups, there is a stable trend. Results for the period between 1996 and 2006 reveal that the Brazilian system is neutral for all cohorts. Therefore, we found out that social security systems are not an effective mechanism for income redistribution, as predicted by previous studies.