Página 1 dos resultados de 21221 itens digitais encontrados em 0.046 segundos

Simulation-based smoothing and filtering in factor stochastic volatility models : two econometric applications

Lopes, Hedibert Freitas
Fonte: Escola de Pós-Graduação em Economia da FGV Publicador: Escola de Pós-Graduação em Economia da FGV
Tipo: Relatório
Português
Relevância na Pesquisa
56.722515%
In this article we use factor models to describe a certain class of covariance structure for financiaI time series models. More specifical1y, we concentrate on situations where the factor variances are modeled by a multivariate stochastic volatility structure. We build on previous work by allowing the factor loadings, in the factor mo deI structure, to have a time-varying structure and to capture changes in asset weights over time motivated by applications with multi pIe time series of daily exchange rates. We explore and discuss potential extensions to the models exposed here in the prediction area. This discussion leads to open issues on real time implementation and natural model comparisons.

Using irregularly spaced returns to estimate multi-factor models : application to Brazilian equity data

Souza, Leonardo R.
Fonte: Escola de Pós-Graduação em Economia da FGV Publicador: Escola de Pós-Graduação em Economia da FGV
Tipo: Trabalho em Andamento
Português
Relevância na Pesquisa
56.441313%
Multi-factor models constitute a use fui tool to explain cross-sectional covariance in equities retums. We propose in this paper the use of irregularly spaced returns in the multi-factor model estimation and provide an empirical example with the 389 most liquid equities in the Brazilian Market. The market index shows itself significant to explain equity returns while the US$/Brazilian Real exchange rate and the Brazilian standard interest rate does not. This example shows the usefulness of the estimation method in further using the model to fill in missing values and to provide intervaI forecasts.

Factor-Augmented Error Correction Models

BANERJEE, Anindya; MARCELLINO, Massimiliano
Fonte: European University Institute Publicador: European University Institute
Tipo: Trabalho em Andamento Formato: application/pdf; digital
Português
Relevância na Pesquisa
46.6948%
This paper brings together several important strands of the econometrics literature: errorcorrection, cointegration and dynamic factor models. It introduces the Factor-augmented Error Correction Model (FECM), where the factors estimated from a large set of variables in levels are jointly modelled with a few key economic variables of interest. With respect to the standard ECM, the FECM protects, at least in part, from omitted variable bias and the dependence of cointegration analysis on the specific limited set of variables under analysis. It may also be in some cases a refinement of the standard Dynamic Factor Model (DFM), since it allows us to include the error correction terms into the equations, and by allowing for cointegration prevent the errors from being non-invertible moving average processes. In addition, the FECM is a natural generalization of factor augmented VARs (FAVAR) considered by Bernanke, Boivin and Eliasz (2005) inter alia, which are specified in first differences and are therefore misspecified in the presence of cointegration. The FECM has a vast range of applicability. A set of Monte Carlo experiments and two detailed empirical examples highlight its merits in finite samples relative to standard ECM and FAVAR models. The analysis is conducted primarily within an in-sample framework...

Dynamic factor models in estimation and forecasting

BYSTROV, Victor
Fonte: Instituto Universitário Europeu Publicador: Instituto Universitário Europeu
Tipo: Tese de Doutorado Formato: application/pdf; digital
Português
Relevância na Pesquisa
46.6948%
This thesis addresses the issue of the relative performance of dynamic factor models in finite samples in the presence of structural breaks. It extends an existing literature by considering new data sets and evaluating finite sample properties of dynamic factor models and factor-augmented VARs and VECMs in Monte Carlo exercises. Chapter 1 Forecasting Emerging Market Indicators: Brazil and Russia Chapter 2 Co-Breaking and Forecasting Performance of Factor Models Chapter 3 Factor Augmented Error Correction Models; Defence date: 6 March 2008; Examining Board: Supervisor: Anindya Banerjee Second reader: Helmut Luetkepohl; First made available online 2 June 2015.

Forecasting with Factor-augmented Error Correction Models

MASTEN, Igor; BANERJEE, Anindya; MARCELLINO, Massimiliano
Fonte: Instituto Universitário Europeu Publicador: Instituto Universitário Europeu
Tipo: Trabalho em Andamento Formato: application/pdf; digital
Português
Relevância na Pesquisa
56.694663%
As a generalization of the factor-augmented VAR (FAVAR) and of the Error Correction Model (ECM), Banerjee and Marcellino (2009) introduced the Factor-augmented Error Correction Model (FECM). The FECM combines error-correction, cointegration and dynamic factor models, and has several conceptual advantages over standard ECM and FAVAR models. In particular, it uses a larger dataset compared to the ECM and incorporates the long-run information lacking from the FAVAR because of the latter’s specification in differences. In this paper we examine the forecasting performance of the FECM by means of an analytical example, Monte Carlo simulations and several empirical applications. We show that relative to the FAVAR, FECM generally offers a higher forecasting precision and in general marks a very useful step forward for forecasting with large datasets.

Forecasting Large Datasets with Bayesian Reduced Rank Multivariate Models

CARRIERO, Andrea; KAPETANIOS, George; MARCELLINO, Massimiliano
Fonte: Instituto Universitário Europeu Publicador: Instituto Universitário Europeu
Tipo: Trabalho em Andamento Formato: application/pdf; digital
Português
Relevância na Pesquisa
56.50667%
The paper addresses the issue of forecasting a large set of variables using multivariate models. In particular, we propose three alternative reduced rank forecasting models and compare their predictive performance for US time series with the most promising existing alternatives, namely, factor models, large scale Bayesian VARs, and multivariate boosting. Speci.cally, we focus on classical reduced rank regression, a two-step procedure that applies, in turn, shrinkage and reduced rank restrictions, and the reduced rank Bayesian VAR of Geweke (1996). We find that using shrinkage and rank reduction in combination rather than separately improves substantially the accuracy of forecasts, both when the whole set of variables is to be forecast, and for key variables such as industrial production growth, inflation, and the federal funds rate. The robustness of this finding is confirmed by a Monte Carlo experiment based on bootstrapped data. We also provide a consistency result for the reduced rank regression valid when the dimension of the system tends to infinity, which opens the ground to use large scale reduced rank models for empirical analysis.

Factor Augmented Error Correction Models

BANERJEE, Anindya; MARCELLINO, Massimiliano
Fonte: Instituto Universitário Europeu Publicador: Instituto Universitário Europeu
Tipo: Parte de Livro
Português
Relevância na Pesquisa
56.743574%
This chapter brings together several important strands of the econometrics literature: error-correction, cointegration, and dynamic factor models. It introduces the Factor-augmented Error Correction Model (FECM), where the factors estimated from a large set of variables in levels are jointly modelled with a few key economic variables of interest. With respect to the standard ECM, the FECM protects, at least in part, from omitted variable bias and the dependence of cointegration analysis on the specific limited set of variables under analysis. It may also be in some cases a refinement of the standard Dynamic Factor Model since it allows the inclusion of error correction terms into the equations, and by allowing for cointegration prevents the errors from being non-invertible moving average processes. In addition, the FECM is a natural generalization of factor augmented VARs (FAVAR) considered by Bernanke, Boivin and Eliasz (2005) inter alia, which are specified in first differences and are therefore misspecified in the presence of cointegration. The FECM has a vast range of applicability. A set of Monte Carlo experiments and two detailed empirical examples highlight its merits in finite samples relative to standard ECM and FAVAR models. The analysis is conducted primarily within an in-sample framework...

Forecasting with factor-augmented error correction models

BANERJEE, Anindya; MARCELLINO, Massimiliano; MASTEN, Igor
Fonte: Elsevier Science Bv Publicador: Elsevier Science Bv
Tipo: Artigo de Revista Científica
Português
Relevância na Pesquisa
56.722515%
As a generalization of the factor-augmented VAR (FAVAR) and of the Error Correction Model (ECM), Banerjee and Marcellino (2009) introduced the Factor-augmented Error Correction Model (FECM). The FECM combines error-correction, cointegration and dynamic factor models, and has several conceptual advantages over the standard ECM and FAVAR models. In particular, it uses a larger dataset than the ECM and incorporates the long-run information which the FAVAR is missing because of its specification in differences. In this paper, we examine the forecasting performance of the FECM by means of an analytical example, Monte Carlo simulations and several empirical applications. We show that FECM generally offers a higher forecasting precision relative to the FAVAR, and marks a useful step forward for forecasting with large datasets.; This article is based on EUI RSCAS WP 2009/32

Detecting big structural breaks in large factor models

Chen, Liang; Dolado, Juan José; Gonzalo, Jesús
Fonte: Universidade Carlos III de Madrid Publicador: Universidade Carlos III de Madrid
Tipo: info:eu-repo/semantics/draft; info:eu-repo/semantics/workingPaper Formato: application/pdf
Publicado em /12/2011 Português
Relevância na Pesquisa
56.57193%
Time invariance of factor loadings is a standard assumption in the analysis of large factor models. Yet, this assumption may be restrictive unless parameter shifts are mild (i.e., local to zero). In this paper we develop a new testing procedure to detect big breaks in these loadings at either known or unknown dates. It relies upon testing for parameter breaks in a regression of the first of the r¯ factors estimated by PCA on the remaining r¯ − 1 factors, where r¯ is chosen according to Bai and Ng’s (2002) information criteria. The test fares well in terms of power relative to other recently proposed tests on this issue, and can be easily implemented to avoid forecasting failures in standard factor-augmented (FAR, FAVAR) models where the number of factors is a priori imposed on the basis of theoretical considerations.

Essays in high dimensional factor models

Chen, Liang
Fonte: Universidade Carlos III de Madrid Publicador: Universidade Carlos III de Madrid
Tipo: Tese de Doutorado
Português
Relevância na Pesquisa
46.85154%
My PhD thesis consists of three chapters on high dimensional factor models and their applications. In Chapter 1, I study how to test for structural breaks in large factor models. Time invariance of factor loadings is a standard assumption in the analysis of large factor models. Yet, this assumption may be restrictive unless parameter shifts are mild. In this chapter we develop a new testing procedure to detect big breaks in these loadings at either known or unknown dates. The test fares well in terms of power relative to other recently proposed tests on this issue, and can be easily implemented to avoid forecasting failures in standard factor-augmented models where the number of factors is a priori imposed on the basis of theoretical considerations. Despite their growing popularity, factor models have been often criticized for lack of identification of the factors. In Chapter 2, I try to identify the orthogonal factors estimated using principal component by associating them to a relevant subset of observed variables. I first propose a selection procedure to choose such a subset, and then test the hypothesis that true factors are exact linear combinations of the selected variables. The good performance of my method in finite samples and its advantages relative to the other available procedures are confirmed through simulations. Empirical applications include the identification of the underlying risk factors in large dataset of stock and portfolio returns...

Latent Factor Models and Analyses for Operator Response Times

Gaver, Donald Paul; O'Muircheartaigh, I. G.
Fonte: Monterey, California. Naval Postgraduate School Publicador: Monterey, California. Naval Postgraduate School
Tipo: Relatório
Português
Relevância na Pesquisa
56.599175%
Two models are presented for the response times of different operators to different tasks where response is initiated by one or more cues provided by the system. One model for the log-response times is a mixed or latent factor model with unequal case fixed effects and variances. The other model for the log-response times is a non-Gaussian log-extreme-value model. Procedures for estimating the parameters by maximum likelihood are presented. The models are used to analyze response time data from simulator experiments involving nuclear power plant operators performing certain safety-related tasks. The findings of the models are critiqued and applications to risk analysis are sketched. Keyword: Extreme-value distribution; Weibull distribution. (KR); Naval Postgraduate School Research Council Research Program; http://archive.org/details/latentfactormode00gave; O&MN, Direct Funding; NA

Factor-MIDAS for Now- and Forecasting with Ragged-Edge Data: A Model Comparison for German GDP

MARCELLINO, Massimiliano; SCHUMACHER, Christian
Fonte: European University Institute Publicador: European University Institute
Tipo: Trabalho em Andamento Formato: application/pdf
Português
Relevância na Pesquisa
56.82196%
This paper compares different ways to estimate the current state of the economy using factor models that can handle unbalanced datasets. Due to the different release lags of business cycle indicators, data unbalancedness often emerges at the end of multivariate samples, which is some- times referred to as the ragged edge of the data. Using a large monthly dataset of the German economy, we compare the performance of different factor models in the presence of the ragged edge: static and dynamic principal components based on realigned data, the Expectation-Maximisation (EM) algorithm and the Kalman smoother in a state-space model context. The monthly factors are used to estimate current quarter GDP, called the nowcast , using different versions of what we call factor-based mixed-data sampling (Factor-MIDAS) approaches. We compare all possible combinations of factor estimation methods and Factor-MIDAS projections with respect to now- cast performance. Additionally, we compare the performance of the nowcast factor models with the performance of quarterly factor models based on time-aggregated and thus balanced data, which neglect the most timely observations of business cycle indicators at the end of the sample. Our empirical ndings show that the factor estimation methods don't differ much with respect to nowcasting accuracy. Concerning the projections...

Dynamic Factor Models in Macro-finance

SCHERRER, David
Fonte: Instituto Universitário Europeu Publicador: Instituto Universitário Europeu
Tipo: Tese de Doutorado
Português
Relevância na Pesquisa
56.590015%
Defence date: 7 June 2011; Examining Board: Professor Richard Spady, Johns Hopkins University (External Supervisor), Professor Helmut Lütkepohl, European University Institute, Professor Marco Lippi, Università La Sapienza, Rome, Professor Emanuel Moench, Federal Reserve Bank of New York; Macroeconomic concepts such as inflation and real economic activity are not directly observed. Researchers often use factor models in order to measure these unobserved concepts. The underlying view is that a small number of factors exist which represent the concept and drive many related variables. Consequently the U.S. economy is often modeled as an affine function of some factors. If indeed there is such a factor structure for the U.S. economy then it can be represented by a generalized dynamic factor model (GDFM). In the first chapter I describe and summarize the literature on GDFMs. In the second chapter I investigate the interactions and mutually independent dynamics of changes in inflation and real growth by applying the GDFM to a block of real growth variables a block of inflation variables and to their joint panel. In this manner an empirical decomposition of the U.S. economy is obtained and this allows the reconcilitaion of forward and backward looking Phillips curves. In the third chapter I build and study a discrete time generalized dynamic affine term structure model. This is characterized by three main features that are conceptually important for a ne yield curve models. I allow: (a) for state vector dynamics beyond Markovian types (b) that all yields may contain an idiosyncratic component to reflect measurement-errors in the data and (c) that idiosyncratic components may be cross sectional as well as time-serial correlated. It is possible to directly compare this model with the version that is restricted by Du e-Kan's no-arbitrage conditions. Chapter four addresses whether or not changes in yields can be explained by changes to the latent dynamic factors which underlie the macroeconomic concepts of inflation and real growth. As such I contribute to the debate about whether or not monetary policy should react to real activity measures.

Latent Variable Models for Stochastic Discount Factors.

GARCIA, René; RENAULT, Éric
Fonte: Université de Montréal Publicador: Université de Montréal
Tipo: Artigo de Revista Científica Formato: 712174 bytes; application/pdf
Português
Relevância na Pesquisa
56.75603%
Latent variable models in finance originate both from asset pricing theory and time series analysis. These two strands of literature appeal to two different concepts of latent structures, which are both useful to reduce the dimension of a statistical model specified for a multivariate time series of asset prices. In the CAPM or APT beta pricing models, the dimension reduction is cross-sectional in nature, while in time-series state-space models, dimension is reduced longitudinally by assuming conditional independence between consecutive returns, given a small number of state variables. In this paper, we use the concept of Stochastic Discount Factor (SDF) or pricing kernel as a unifying principle to integrate these two concepts of latent variables. Beta pricing relations amount to characterize the factors as a basis of a vectorial space for the SDF. The coefficients of the SDF with respect to the factors are specified as deterministic functions of some state variables which summarize their dynamics. In beta pricing models, it is often said that only the factorial risk is compensated since the remaining idiosyncratic risk is diversifiable. Implicitly, this argument can be interpreted as a conditional cross-sectional factor structure...

Factor models, VARMA processes and parameter instability with applications in macroeconomics

Stevanovic, Dalibor
Fonte: Université de Montréal Publicador: Université de Montréal
Tipo: Thèse ou Mémoire numérique / Electronic Thesis or Dissertation
Português
Relevância na Pesquisa
46.89435%
Avec les avancements de la technologie de l'information, les données temporelles économiques et financières sont de plus en plus disponibles. Par contre, si les techniques standard de l'analyse des séries temporelles sont utilisées, une grande quantité d'information est accompagnée du problème de dimensionnalité. Puisque la majorité des séries d'intérêt sont hautement corrélées, leur dimension peut être réduite en utilisant l'analyse factorielle. Cette technique est de plus en plus populaire en sciences économiques depuis les années 90. Étant donnée la disponibilité des données et des avancements computationnels, plusieurs nouvelles questions se posent. Quels sont les effets et la transmission des chocs structurels dans un environnement riche en données? Est-ce que l'information contenue dans un grand ensemble d'indicateurs économiques peut aider à mieux identifier les chocs de politique monétaire, à l'égard des problèmes rencontrés dans les applications utilisant des modèles standards? Peut-on identifier les chocs financiers et mesurer leurs effets sur l'économie réelle? Peut-on améliorer la méthode factorielle existante et y incorporer une autre technique de réduction de dimension comme l'analyse VARMA? Est-ce que cela produit de meilleures prévisions des grands agrégats macroéconomiques et aide au niveau de l'analyse par fonctions de réponse impulsionnelles? Finalement...

The Earnings/Price Risk Factor in Capital Asset Pricing Models

Noda,Rafael Falcão; Martelanc,Roy; Kayo,Eduardo Kazuo
Fonte: Universidade de São Paulo, Faculdade de Economia, Administração e Contabilidade, Departamento de Contabilidade e Atuária Publicador: Universidade de São Paulo, Faculdade de Economia, Administração e Contabilidade, Departamento de Contabilidade e Atuária
Tipo: Artigo de Revista Científica Formato: text/html
Publicado em 01/01/2015 Português
Relevância na Pesquisa
56.781006%
This article integrates the ideas from two major lines of research on cost of equity and asset pricing: multi-factor models and ex ante accounting models. The earnings/price ratio is used as a proxy for the ex ante cost of equity, in order to explain realized returns of Brazilian companies within the period from 1995 to 2013. The initial finding was that stocks with high (low) earnings/price ratios have higher (lower) risk-adjusted realized returns, already controlled by the capital asset pricing model's beta. The results show that selecting stocks based on high earnings/price ratios has led to significantly higher risk-adjusted returns in the Brazilian market, with average abnormal returns close to 1.3% per month. We design asset pricing models including an earnings/price risk factor, i.e. high earnings minus low earnings, based on the Fama and French three-factor model. We conclude that such a risk factor is significant to explain returns on portfolios, even when controlled by size and market/book ratios. Models including the high earnings minus low earnings risk factor were better to explain stock returns in Brazil when compared to the capital asset pricing model and to the Fama and French three-factor model, having the lowest number of significant intercepts. These findings may be due to the impact of historically high inflation rates...

Infinite Dimensional VARs and Factor Models

Chudik, Alexander; Pesaran, M. Hashem
Fonte: Faculty of Economics, University of Cambridge, UK Publicador: Faculty of Economics, University of Cambridge, UK
Tipo: Trabalho em Andamento
Português
Relevância na Pesquisa
56.590015%
This paper introduces a novel approach for dealing with the 'curse of dimensionality' in the case of large linear dynamic systems. Restrictions on the coefficients of an unrestricted VAR are proposed that are binding only in a limit as the number of endogenous variables tends to infinity. It is shown that under such restrictions, an infinite-dimensional VAR (or IVAR) can be arbitrarily well characterized by a large number of finite-dimensional models in the spirit of the global VAR model proposed in Pesaran et al. (JBES, 2004). The paper also considers IVAR models with dominant individual units and shows that this will lead to a dynamic factor model with the dominant unit acting as the factor. The problems of estimation and inference in a stationary IVAR with unknown number of unobserved common factors are also investigated. A cross section augmented least squares estimator is proposed and its asymptotic distribution is derived. Satisfactory small sample properties are documented by Monte Carlo experiments. An empirical application to modelling of real GDP growth and investment-output ratios provides an illustration of the proposed approach. Considerable heterogeneities across countries and significant presence of dominant effects are found. The results also suggest that increase in investment as a share of GDP predict higher growth rate of GDP per capita for non-negligible fraction of countries and vice versa.

Factor Models to Describe Linear and Non-linear Structure in High Dimensional Gene Expression Data

Mayrink, Vinicius Diniz
Fonte: Universidade Duke Publicador: Universidade Duke
Tipo: Dissertação
Publicado em //2011 Português
Relevância na Pesquisa
56.693965%

An important problem in the analysis of gene expression data is the identification of groups of features that are coherently expressed. For example, one often wishes to know whether a group of genes, clustered because of correlation in one data set, is still highly co-expressed in another data set. For some microarray platforms there are many, relatively short, probes for each gene of interest. In this case, it is possible that a given probe is not measuring its targeted transcript, but rather a different gene with a similar region (called cross-hybridization). Similarly, the incorrect mapping of short nucleotide sequences to a target gene is a common issue related to the young technology producing RNA-Seq data. The expression pattern across samples is a valuable source of information, which can be used to address distinct problems through the application of factor models. Our first study is focused on the identification of the presence/absence status of a gene in a sample. We compare our factor model to state-of-the-art detection methods; the results suggest superior performance of the factor analysis for detecting transcripts. In the second study, we apply factor models to investigate gene modules (groups of coherently expressed genes). Variation in the number of copies of regions of the genome is a well known and important feature of most cancers. Copy number alteration is detected for a group of genes in breast cancer; our goal is to examine this abnormality in the same chromosomal region for other types of tumors (Ovarian...

Bayesian Semi-parametric Factor Models

Bhattacharya, Anirban
Fonte: Universidade Duke Publicador: Universidade Duke
Tipo: Dissertação
Publicado em //2012 Português
Relevância na Pesquisa
56.838354%

Identifying a lower-dimensional latent space for representation of high-dimensional observations is of significant importance in numerous biomedical and machine learning applications. In many such applications, it is now routine to collect data where the dimensionality of the outcomes is comparable or even larger than the number of available observations. Motivated in particular by the problem of predicting the risk of impending diseases from massive gene expression and single nucleotide polymorphism profiles, this dissertation focuses on building parsimonious models and computational schemes for high-dimensional continuous and unordered categorical data, while also studying theoretical properties of the proposed methods. Sparse factor modeling is fast becoming a standard tool for parsimonious modeling of such massive dimensional data and the content of this thesis is specifically directed towards methodological and theoretical developments in Bayesian sparse factor models.

The first three chapters of the thesis studies sparse factor models for high-dimensional continuous data. A class of shrinkage priors on factor loadings are introduced with attractive computational properties, with operating characteristics explored through a number of simulated and real data examples. In spite of the methodological advances over the past decade...

On Bayesian Analyses of Functional Regression, Correlated Functional Data and Non-homogeneous Computer Models

Montagna, Silvia
Fonte: Universidade Duke Publicador: Universidade Duke
Tipo: Dissertação
Publicado em //2013 Português
Relevância na Pesquisa
56.533086%

Current frontiers in complex stochastic modeling of high-dimensional processes include major emphases on so-called functional data: problems in which the data are snapshots of curves and surfaces representing fundamentally important scientific quantities. This thesis explores new Bayesian methodologies for functional data analysis.

The first part of the thesis places emphasis on the role of factor models in functional data analysis. Data reduction becomes mandatory when dealing with such high-dimensional data, more so when data are available on a large number of individuals. In Chapter 2 we present a novel Bayesian framework which employs a latent factor construction to represent each variable by a low dimensional summary. Further, we explore the important issue of modeling and analyzing the relationship of functional data with other covariate and outcome variables simultaneously measured on the same subjects.

The second part of the thesis is concerned with the analysis of circadian data. The focus is on the identification of circadian genes that is, genes whose expression levels appear to be rhythmic through time with a period of approximately 24 hours. While addressing this goal, most of the current literature does not account for the potential dependence across genes. In Chapter 4...