Dynamic Factor Models and Forecasting Finnish Macroeconomic Variables

In recent years, thanks to developments in information technology, large-dimensional datasets have been increasingly available. Researchers now have access to thousands of economic series and the information contained in them can be used to create accurate forecasts and to test economic theories. To exploit this large amount of information, researchers and policymakers need an appropriate econometric model.Usual time series models, vector autoregression for example, cannot incorporate more than a few variables. There are two ways to solve this problem: use variable selection procedures or gather the information contained in the series to create an index model. This thesis focuses on one of the most widespread index model, the dynamic factor model (the theory behind this model, based on previous literature, is the core of the first part of this study), and its use in forecasting Finnish macroeconomic indicators (which is the focus of the second part of the thesis). In particular, I forecast economic activity indicators (e.g. GDP) and price indicators (e.g. consumer price index), from 3 large Finnish datasets. The first dataset contains a large series of aggregated data obtained from the Statistics Finland database. The second dataset is composed by economic indicators from Bank of Finland. The last dataset is formed by disaggregated data from Statistic Finland, which I call micro dataset. The forecasts are computed following a two steps procedure: in the first step I estimate a set of common factors from the original dataset. The second step consists in formulating forecasting equations including the factors extracted previously. The predictions are evaluated using relative mean squared forecast error, where the benchmark model is a univariate autoregressive model. The results are dataset-dependent. The forecasts based on factor models are very accurate for the first dataset (the Statistics Finland one), while they are considerably worse for the Bank of Finland dataset. The forecasts derived from the micro dataset are still good, but less accurate than the ones obtained in the first case. This work leads to multiple research developments. The results here obtained can be replicated for longer datasets. The non-aggregated data can be represented in an even more disaggregated form (firm level). Finally, the use of the micro data, one of the major contributions of this thesis, can be useful in the imputation of missing values and the creation of flash estimates of macroeconomic indicator (nowcasting).

ARE OECD MACROECONOMIC VARIABLES TREND STATIONARY? EVIDENCE FROM PANEL STATIONARITY TESTS ALLOWING FOR A STRUCTURAL BREAK AND CROSS-SECTIONAL DEPENDENCE

This article applies the panel stationarity test with a break proposed by Hadri and Rao (2008) to examine whether 14 macroeconomic variables of OECD countries can be best represented as random walk or stationary fluctuations around a deterministic trend. In contrast to previous studies, based essentially on visual inspection of the break type or just applying the most general break model, we use a model selection procedure based on BIC. We do this for each time series so that heterogeneous break models are allowed for in the panel. Our results suggest, overwhelmingly, that if we account for a structural break, cross-sectional dependence and choose the break models to be congruent with the data, then the null of stationarity cannot be rejected for all the 14 macroeconomic variables examined in this article. This is in sharp contrast with the results obtained by Hurlin (2004), using the same data but a different methodology.

Do US macroeconomic forecasters exaggerate their differences?

Application of the Bernhardt et al. (Journal of Financial Economics 2006; 80(3): 657–675) test of herding to the calendar-year annual output growth and inflation forecasts suggests forecasters tend to exaggerate their differences, except at the shortest horizon, when they tend to herd. We consider whether these types of behaviour can help to explain the puzzle that professional forecasters sometimes make point predictions and histogram forecasts which are mutually inconsistent.

Price setting and macroeconomic variables: evidence from Brazilian CPI

Esta tese investiga as estratégias de precificação em ambientes macroeconômicos distintos, utilizando uma base de dados única para o IPC da Fundação Getulio Vargas. A base de dados primária consiste em um painel de dados individuais para bens e serviços representando 100% do IPC para o período de 1996 a 2008. Durante este período, diversos eventos produziram uma variabilidade macroeconômica substancial no Brasil: duas crises em países emergentes, uma mudança de regime cambial e monetário, racionamento de energia, uma crise de expectativas eleitorais e um processo de desinflação. Como consequência, a inflação, a incerteza macroeconômica, a taxa de câmbio e o produto exibiram uma variação considerável no período. No primeiro capítulo, nós descrevemos a base de dados e apresentamos as principais estatísticas de price-setting para o Brasil. Em seguida, nos capítulos 2 e 3, nos construímos as séries de tempo destas estatísticas e das estatísticas de promoções, e as relacionamos com as variáveis macroeconômicas utilizando análises de regressões. Os resultados indicam que há uma relação substancial entre as estatísticas de price-setting e o ambiente macroeconômico para a economia brasileira.

Brazil: how macroeconomic variables affect consumer confidence

Includes bibliography

Efeitos de política fiscal nos EUA em um modelo FAVAR

O objetivo desse artigo é analisar o impacto da política fiscal sobre diversas variáveis macroeconômicas dos EUA. A metodologia do trabalho empírico baseia-se em um modelo VAR estrutural que incorpora fatores latentes (FAVAR) e para o qual desenvolve-se um esquema de identificação específico. Visto que os fatores são estimados por principal components, estes aproximam-se muito das séries observadas de produção industrial e taxa de juros. Como será visto, este resultado é de fundamental importância para a hipótese de identificação e a escolha dos instrumentos do modelo VAR. Por meio das funções de resposta ao impulso analisa-se os efeitos de um aumento do gasto do governo sobre variáveis de produto e consumo e, por sua vez, corroborando a hipótese de que tanto o PIB quanto as despesas de consumo das famílias aumentam depois desse choque exógeno. Em particular esse efeito sobre o consumo também é verificado quando separamos os indivíduos em várias classes de acordo com renda. Olhando cuidadosamente no entanto pode-se perceber que um aumento no gasto público possui mais impacto sobre os consumidores de renda mais baixa. Ou seja, é provável que por estarem sujeitas a restrições de crédito, as classes mais baixas tem mais dificuldade em suavizar o consumo após um choque agregado.

The influence of macroeconomic factors on primary issues in the Brazilian market

This research analyses the influence of the macroeconomic factors on the primary issue of stocks and debentures in the Brazilian market. Previous studies have agreed on the importance of aspects of the economic situation on a company’s capital structure, but have not established a relationship between the macroeconomic variables and the level of aggregate debt; we can mention Procianoy and Caselani (1997) and Terra (2003) as examples of this. According to Leal (2000), the limitations of the Brazilian capital market suggest that management takes advantage of moments of euphoria in the market - whether caused by a reduction in the rate of interest or by the return being offered by the equity market - to raise funds at rates that are more advantageous to the company. This characterizes the first evidence we have of opportunistic behavior influencing a company’s financing decisions. Eid Jr. (1996) provides us with the first evidence of this opportunistic behavior in his research in which 47% of those interviewed said that they chose fund sources that are economically more advantageous.

Geographic Deregulation and Commercial Bank Performance in US State Banking Markets

This paper examines the effects of geographical deregulation on commercial bank performance across states. We reach some general conclusions. First, the process of deregulation on an intrastate and interstate basis generally improves bank profitability and performance. Second, the macroeconomic variables -- the unemployment rate and real personal income per capita -- and the average interest rate affect bank performance as much, or more, than the process of deregulation. Finally, while deregulation toward full interstate banking and branching may produce more efficient banks and a healthier banking system, we find mixed results on this issue.

Essays on Labor Market Frictions, Technological Change and Macroeconomic Fluctuations

The dissertation consists of an introductory chapter and three essays that apply search-matching theory to study the interaction of labor market frictions, technological change and macroeconomic fluctuations. The first essay studies the impact of capital-embodied growth on equilibrium unemployment by extending a vintage capital/search model to incorporate vintage human capital. In addition to the capital obsolescence (or creative destruction) effect that tends to raise unemployment, vintage human capital introduces a skill obsolescence effect of faster growth that has the opposite sign. Faster skill obsolescence reduces the value of unemployment, hence wages and leads to more job creation and less job destruction, unambiguously reducing unemployment. The second essay studies the effect of skill biased technological change on skill mismatch and the allocation of workers and firms in the labor market. By allowing workers to invest in education, we extend a matching model with two-sided heterogeneity to incorporate an endogenous distribution of high and low skill workers. We consider various possibilities for the cost of acquiring skills and show that while unemployment increases in most scenarios, the effect on the distribution of vacancy and worker types varies according to the structure of skill costs. When the model is extended to incorporate endogenous labor market participation, we show that the unemployment rate becomes less informative of the state of the labor market as the participation margin absorbs employment effects. The third essay studies the effects of labor taxes on equilibrium labor market outcomes and macroeconomic dynamics in a New Keynesian model with matching frictions. Three policy instruments are considered: a marginal tax and a tax subsidy to produce tax progression schemes, and a replacement ratio to account for variability in outside options. In equilibrium, the marginal tax rate and replacement ratio dampen economic activity whereas tax subsidies boost the economy. The marginal tax rate and replacement ratio amplify shock responses whereas employment subsidies weaken them. The tax instruments affect the degree to which the wage absorbs shocks. We show that increasing tax progression when taxation is initially progressive is harmful for steady state employment and output, and amplifies the sensitivity of macroeconomic variables to shocks. When taxation is initially proportional, increasing progression is beneficial for output and employment and dampens shock responses.

Data Revisions in the Estimation of DSGE Models

Revisions of US macroeconomic data are not white-noise. They are persistent, correlated with real-time data, and with high variability (around 80% of volatility observed in US real-time data). Their business cycle effects are examined in an estimated DSGE model extended with both real-time and final data. After implementing a Bayesian estimation approach, the role of both habit formation and price indexation fall significantly in the extended model. The results show how revision shocks of both output and inflation are expansionary because they occur when real-time published data are too low and the Fed reacts by cutting interest rates. Consumption revisions, by contrast, are countercyclical as consumption habits mirror the observed reduction in real-time consumption. In turn, revisions of the three variables explain 9.3% of changes of output in its long-run variance decomposition.

Macroeconomic effects of monetary policy in the Eurozone

To study the macroeconomic effects of unconventional monetary policy across the different countries of the eurozone, I develop an identification scheme to disentangle conventional from non-conventional policy shocks, using futures contracts on overnight interest rates and the size of the European Central Bank balance sheet. Setting these shocks as endogenous variables in a structural vector autoregressive (SVAR) model, along with the CPI and the employment rate, estimated impulse response functions of policy to macroeconomic variables are studied. I find that unconventional policy shocks generated mixed effects in inflation but had a positive impact on employment, with the exception of Portugal, Spain, Greece and Italy where the employment response is close to zero or negative. The heterogeneity that characterizes the responses shows that the monetary policy measures taken in recent years were not sufficient to stabilize the economies of the eurozone countries under more severe economic conditions.