[pt] Mavroeidis (2005) alertou que equações de equlíbrio motivadas por modelos macroeconômicos com expectatvas racionais poderiam ser fracamente identificados devido ao uso de instrumentos fracos. Eu argumento que, embora tais preocupações sejam legítimas, elas não são empiricamente graves, contanto que instrumentos sejam devidamente selecionados. Eu utilizo um modelo DSGE estimado de média escala como laboratório para avaliar estimação uniequacional de condições de equilíbrio macroeconômicas. Apresento estimadores baseados no LASSO que selecionam instrumentos e tem boa performance em amostra finita, que argumento funcionam melhor em relações que incluem termos de expectativa, como a Curva de Phillips Novo Keynesiana. Por último, faço uma aplicação empírica para a Curva de Phillips da economia dos Estados-Unidos e as estimativas validam um componente de expectativa predominante. / [en] Mavroeidis (2005) alerted that equilibrium conditions from typical rational expectations macroeconomic models could be weakly identified due to the use of poor instruments. I argue that, although such concerns are legitimate, they are not empirically severe, provided instruments are properly selected. I use an estimated medium scale DSGE model as a laboratory to assess single-equation estimation of macroeconomic equilibrium conditions. I present LASSO-based estimators to select instruments that perform well in finite samples, which I argue have a better chance of performing for forward-looking relationships, such as the New Keynesian Phillips Curve. Finally, I provide an empirical application of the estimators for the US economy s Phillips Curve and show that it validates a dominant forward looking behavior.

Thèse numérisée par la Division de la gestion de documents et des archives de l'Université de Montréal.

Thèse numérisée par la Division de la gestion de documents et des archives de l'Université de Montréal

New keynesian Phillips curve (NKPC) has become a central model to study the relation between inflation and real economic activity, notably in the framework of optimal monetary policy design. However, some recent evidence suggests that empirical data are usually at odds with the underlying theory. The model due to its inherent structure represents a statistical challenge in its own right. Since Galí and Gertler (1999) published their seminal paper introducing estimation via GMM techniques, they have triggered a heated debate on its empirical relevance. Their approach has been heavily criticised by later authors, mainly on the grounds of questionable behaviour of GMM estimator in the NKPC context and/or its small sample properties. The common criticism includes sensitivity to the choice of instrument set, weak identification and small sample bias. In this thesis I propose a new estimation strategy that provides a remedy to above mentioned shortcomings and allows to obtain reliable estimates. The procedure exploits recent advances in GMM theory as well as in other fields of statistics, in particular in the area of time series factor analysis and bootstrap. The proposed estimation strategy consists of several consecutive steps: first, to reduce a small sample bias resulting from excessive use of instruments I summarize all available information by employing factor analysis and include estimated factors into information set. In the second step I use statistical information criteria to select optimal instruments and eventually I obtain confidence intervals on parameters using bootstrap method. In NKPC context all these methods were used for the first time and can also be used independently. Their combination however provides synergistic effect that helps to improve the properties of estimates and to check the efficiency of given steps. Obtained results suggest that NKPC model can explain Czech inflation dynamics fairly well and provide some support for underlying theory. Among other things the results imply that the policy of disinflation may not be as costly with respect to a loss in aggregate product as earlier versions of Phillips curve would indicate. However, finding a good proxy for real economic activity has proved to be a difficult task. In particular we demonstrated that results are conditional on how the measure is calculated, some measures even showed countercyclical behaviour. This issue -- in the thesis discussed only in passing -- is a subject of future research. In addition to the proposed strategy and provided parameter estimates the thesis brings some partial simulation-based findings. Simulations elaborate on earlier literature on naive bootstrap in GMM context and study performance of bootstrap modifications of unit root and KPSS test.

Submitted by Leonardo Salim Saker Chaves (lsalimsaker@gmail.com) on 2015-07-24T19:51:22Z No. of bitstreams: 1 Dissertacao_LeonardoSalim_BMHS.pdf: 661288 bytes, checksum: a89da060d1378be5cf51ff1edc18cfc6 (MD5) / Approved for entry into archive by BRUNA BARROS (bruna.barros@fgv.br) on 2015-07-27T14:42:35Z (GMT) No. of bitstreams: 1 Dissertacao_LeonardoSalim_BMHS.pdf: 661288 bytes, checksum: a89da060d1378be5cf51ff1edc18cfc6 (MD5) / Approved for entry into archive by Maria Almeida (maria.socorro@fgv.br) on 2015-07-30T19:33:12Z (GMT) No. of bitstreams: 1 Dissertacao_LeonardoSalim_BMHS.pdf: 661288 bytes, checksum: a89da060d1378be5cf51ff1edc18cfc6 (MD5) / Made available in DSpace on 2015-07-30T19:33:33Z (GMT). No. of bitstreams: 1 Dissertacao_LeonardoSalim_BMHS.pdf: 661288 bytes, checksum: a89da060d1378be5cf51ff1edc18cfc6 (MD5) Previous issue date: 2015-04-28 / This work studies the hypothesis testing based on generalized method of moments (GMM) estimation given by instruments condition. The importance for the development of Economics lies on the fact that when identi cation is weak, the standard test can be misleading. Therefore, it is made a review of proposed tests to overcome this problem and also present two useful frameworks of study; from Moreira (2002), Moreira and Moreira (2013) and Kleibergen (2005). So, this work conciliate the previous frameworks a way to write the score proposed initially in Kleibergen (2005) using Moreira and Moreira (2013) statistics and presents the optimal score test based on asymptotic theory from Newey and McFadden (1984). Moreover, the study shows the equivalence between the GMM and maximum likelihood estimation to deal with the weak instruments problem. / Esta dissertação se propõe ao estudo de inferência usando estimação por método generalizado dos momentos (GMM) baseado no uso de instrumentos. A motivação para o estudo está no fato de que sob identificação fraca dos parâmetros, a inferência tradicional pode levar a resultados enganosos. Dessa forma, é feita uma revisão dos mais usuais testes para superar tal problema e uma apresentação dos arcabouços propostos por Moreira (2002) e Moreira & Moreira (2013), e Kleibergen (2005). Com isso, o trabalho concilia as estatísticas utilizadas por eles para realizar inferência e reescreve o teste score proposto em Kleibergen (2005) utilizando as estatísticas de Moreira & Moreira (2013), e é obtido usando a teoria assintótica em Newey & McFadden (1984) a estatística do teste score ótimo. Além disso, mostra-se a equivalência entre a abordagem por GMM e a que usa sistema de equações e verossimilhança para abordar o problema de identificação fraca.

[pt] O problema de identificação fraca surge naturalmente em modelos macroeconômicos. Consequentemente, métodos de variáveis instrumentais produzem resultados enigmáticos de forma mais frequente do que seria empiricamente razoável. Neste trabalho, propomos dois novos métodos para tratar destas dificuldades, no que tange a duas das principais equações de modelos macro: a Curva de Phillips Novo-Keynesiana (NKPC) e a Equação de Euler (EE). Sabe-se das dificuldades em se estimar um coeficiente de sensibilidade positivo entre inflação e produto no primeiro caso, e que, mesmo quando se obtém uma estimativa positiva, o nível de rigidez nominal implicado para a economia é incompatível com o que sugerem os micro dados. Nós abordamos essa questão no primeiro capítulo, propondo um modelo de economia multi-setorial com heterogeneidade na fixação de preços entre setores. O método gera coeficientes de sensibilidade positivos e estáveis para diferentes configurações econométricas, assim como níveis de rigidez nominal alinhados com a evidência micro, para a economia como um todo e também para cada setor individualmente. Todas essas estimativas variam em linha com implicações teóricas, quando hipóteses do modelo são alteradas. O foco do segundo capítulo é a estimação da elasticidade de substituição intertemporal (EIS), parâmetro central da EE. Argumentamos como o uso de séries oficiais de consumo – que são estatisticamente tratadas antes de disponibilizadas – distorce estimativas da EIS. Propondo um modelo generalizado para desfiltrar diferentes tipos de séries de consumo disponíveis, – micro e macro, com várias frequências –, demonstramos como a utilização de consumo não filtrado gera estimativas da EIS que são consideravelmente mais estáveis, independente do arcabouço econométrico e da série de consumo usada. Resultados também parecem menos sensíveis à presença de instrumentos fracos, comparativamente a estimações usando séries oficiais. / [en] The weak identification problem arises naturally in macroeconomic models. Consequently, instrumental variables methods produce puzzling results more often than what is empirically plausible. We propose novel methods to address puzzles usually featured in two of the main equations in macro models, namely the New-Keynesian Phillips Curve (NKPC) and the Euler Equation (EE). For the former, difficulties to estimate a positive slope without incurring a degree of stickiness incompatible with the micro evidence are widely known. We address the matter in the first chapter, proposing a richer framework of a multi-sector economy with price-setting heterogeneity. The procedure generates positive and roughly unchanging slope coefficients across econometric settings, as well as degrees of stickiness in line with the micro data, both regarding the entire economy and the cross section of sectors. Importantly, all of these estimates move consistently with implications by theory when modifying the model assumptions. The second chapter focuses on the estimation of the elasticity of intertemporal substitution (EIS), central parameter of the EE in models of dynamic choice. There, we argue that the use of officially reported consumption data – which is usually filtered, smoothed, interpolated, etc – distorts estimates of the EIS. A generalised model to unfilter available consumption data is proposed, suitable for several types of data – macro and micro – at different frequencies. Estimations based on unfiltered consumption produce considerably more stable estimates of the EIS, regardless of the econometric approach and the type of consumption data used. Results also seem less sensitive to the presence of weak instruments, compared to officially reported data.