1. Introduction and Overview. 2. Detecting Influential Observations and Outliers. 3. Detecting and Assessing Collinearity. 4. Applications and Remedies. 5. Research Issues and Directions for Extensions. Bibliography. Author Index. Subject Index.

Regression Diagnostics: Identifying Influential Data and Sources of Collinearity

In this book, Andrew Harvey sets out to provide a unified and comprehensive theory of structural time series models. Unlike the traditional ARIMA models, structural time series models consist explicitly of unobserved components, such as trends and seasonals, which have a direct interpretation. As a result the model selection methodology associated with structural models is much closer to econometric methodology. The link with econometrics is made even closer by the natural way in which the models can be extended to include explanatory variables and to cope with multivariate time series. From the technical point of view, state space models and the Kalman filter play a key role in the statistical treatment of structural time series models. The book includes a detailed treatment of the Kalman filter. This technique was originally developed in control engineering, but is becoming increasingly important in fields such as economics and operations research. This book is concerned primarily with modelling economic and social time series, and with addressing the special problems which the treatment of such series poses. The properties of the models and the methodological techniques used to select them are illustrated with various applications. These range from the modellling of trends and cycles in US macroeconomic time series to to an evaluation of the effects of seat belt legislation in the UK.

Forecasting, Structural Time Series Models and the Kalman Filter

This paper presents a new approach to estimation and inference in panel data models with a multifactor error structure where the unobserved common factors are (possibly) correlated with exogenously given individual-specific regressors, and the factor loadings differ over the cross section units. The basic idea behind the proposed estimation procedure is to filter the individual-specific regressors by means of (weighted) cross-section aggregates such that asymptotically as the cross-section dimension (N) tends to infinity the differential effects of unobserved common factors are eliminated. The estimation procedure has the advantage that it can be computed by OLS applied to an auxiliary regression where the observed regressors are augmented by (weighted) cross sectional averages of the dependent variable and the individual specific regressors. Two different but related problems are addressed: one that concerns the coefficients of the individual-specific regressors, and the other that focusses on the mean of the individual coefficients assumed random. In both cases appropriate estimators, referred to as common correlated effects (CCE) estimators, are proposed and their asymptotic distribution as N with T (the time-series dimension) fixed or as N and T (jointly) are derived under different regularity conditions. One important feature of the proposed CCE mean group (CCEMG) estimator is its invariance to the (unknown but fixed) number of unobserved common factors as N and T (jointly). The small sample properties of the various pooled estimators are investigated by Monte Carlo experiments that confirm the theoretical derivations and show that the pooled estimators have generally satisfactory small sample properties even for relatively small values of N and T.

Estimation and Inference in Large Heterogeneous Panels with a Multifactor Error Structure

Using a Bayesian likelihood approach, we estimate a dynamic stochastic general equilibrium model for the US economy using seven macro-economic time series. The model incorporates many types of real and nominal frictions and seven types of structural shocks. We show that this model is able to compete with Bayesian Vector Autoregression models in out-of-sample prediction. We investigate the relative empirical importance of the various frictions. Finally, using the estimated model we address a number of key issues in business cycle analysis: What are the sources of business cycle fluctuations? Can the model explain the cross-correlation between output and inflation? What are the effects of productivity on hours worked? What are the sources of the "Great Moderation"?

/pdf/shocks-and-frictions-in-us-business-cycles-a-bayesian-dsge-d1f6ynd58q.pdf

Shocks and Frictions in US Business Cycles: A Bayesian DSGE Approach

This paper presents a new approach to estimation and inference in panel data models with a general multifactor error structure. The unobserved factors and the individual-specific errors are allowed to follow arbitrary stationary processes, and the number of unobserved factors need not be estimated. The basic idea is to filter the individual-specific regressors by means of cross-section averages such that asymptotically as the cross-section dimension (N) tends to infinity, the differential effects of unobserved common factors are eliminated. The estimation procedure has the advantage that it can be computed by least squares applied to auxiliary regressions where the observed regressors are augmented with cross-sectional averages of the dependent variable and the individual-specific regressors. A number of estimators (referred to as common correlated effects (CCE) estimators) are proposed and their asymptotic distributions are derived. The small sample properties of mean group and pooled CCE estimators are investigated by Monte Carlo experiments, showing that the CCE estimators have satisfactory small sample properties even under a substantial degree of heterogeneity and dynamics, and for relatively small values of N and T.

/pdf/estimation-and-inference-in-large-heterogeneous-panels-with-tz1flk44ix.pdf

This paper proposes a factor model with infinite dynamics and nonorthogonal idiosyncratic components. The model, which we call the generalized dynamic-factor model, is novel to the literature and generalizes the static approximate factor model of Chamberlain and Rothschild (1983), as well as the exact factor model a la Sargent and Sims (1977). We provide identification conditions, propose an estimator of the common components, prove convergence as both time and cross-sectional size go to infinity at appropriate rates, and present simulation results. We use our model to construct a coincident index for the European Union. Such index is defined as the common component of real GDP within a model including several macroeconomic variables for each European country.

https://pages.stern.nyu.edu/~dbackus/Identification/FHLR_factor_REStat_00.pdf

The Generalized Dynamic-Factor Model: Identification and Estimation

This article proposes a new forecasting method that makes use of information from a large panel of time series. Like earlier methods, our method is based on a dynamic factor model. We argue that our method improves on a standard principal component predictor in that it fully exploits all the dynamic covariance structure of the panel and also weights the variables according to their estimated signal-to-noise ratio. We provide asymptotic results for our optimal forecast estimator and show that in finite samples, our forecast outperforms the standard principal components predictor.

/pdf/the-generalized-dynamic-factor-model-one-sided-estimation-2lzjkhkuui.pdf

The Generalized Dynamic Factor Model one-sided estimation and forecasting

This Paper proposes a new forecasting method that exploits information from a large panel of time series. The method is based on the generalized dynamic factor model proposed in Forni, Hallin, Lippi, and Reichlin (2000), and takes advantage of the information on the dynamic covariance structure of the whole panel. We first use our previous method to obtain an estimation for the covariance matrices of common and idiosyncratic components. The generalized eigenvectors of this couple of matrices are then used to derive a consistent estimate of the optimal forecast, which is constructed as a linear combination of present and past observations only (one-sided filter). This two-step approach solves the end-of-sample problems caused by two-sided filtering (as in our previous work), while retaining the advantages of an estimator based on dynamic information. Both simulation results and an empirical illustration on the forecast of the Euro area industrial production and inflation, based on a panel of 447 monthly time series show very encouraging results.

The Generalized Dynamic Factor Model: One-Sided Estimation and Forecasting

This paper, along with the companion paper Forni, Hallin, Lippi, and Reichlin (2000, Review of Economics and Statistics 82, 540–554), introduces a new model—the generalized dynamic factor model—for the empirical analysis of financial and macroeconomic data sets characterized by a large number of observations both cross section and over time. This model provides a generalization of the static approximate factor model of Chamberlain (1983, Econometrica 51, 1181–1304) and Chamberlain and Rothschild (1983, Econometrica 51, 1305–1324) by allowing serial correlation within and across individual processes and of the dynamic factor model of Sargent and Sims (1977, in C.A. Sims (ed.), New Methods in Business Cycle Research, pp. 45–109) and Geweke (1977, in D.J. Aigner & A.S. Goldberger (eds.), Latent Variables in Socio-Economic Models, pp. 365–383) by allowing for nonorthogonal idiosyncratic terms. Whereas the companion paper concentrates on identification and estimation, here we give a full characterization of the generalized dynamic factor model in terms of observable spectral density matrices, thus laying a firm basis for empirical implementation of the model. Moreover, the common factors are obtained as limits of linear combinations of dynamic principal components. Thus the paper reconciles two seemingly unrelated statistical constructions.

The generalized dynamic factor model: representation theory

This paper develops a method for analysing the dynamics of large cross-sections based on a factor analytic model. We use "law of large numbers" arguments to show that the number of common factors can be determined by a principal components method, the economy-wide shocks can be identified by means of simple structural VAR techniques and that the parameters of the unobserved factor model can be estimated consistently by applying OLS equation by equation. We distinguish between a technological and a non-technological shock. Identification is obtained by minimizing the negative realizations of the technology shock. Empirical results on 4-digit industrial output and productivity for the U.S. economy from 1958 to 1986 show that: (1) at least two economy-wide shocks, both having a long-run effect on sectoral output, are needed to explain the common dynamics; (2) although the technological shock accounts for at least 50% of the aggregate dynamics of output, it cannot by itself explain dynamics at business cycle frequencies; (3) sector-specific shocks explain the main bulk of total variance but generate mainly high frequency dynamics; (4) both the technological and the non-technological component of output show a peak for positive sectoral comovements of output at business cycle frequencies; (5) technological shocks are strongly correlated with the growth rates of the investment in machinery and equipment sectors and their inputs. Many interesting questions about cyclical fluctuations and economic growth can be answered only by studying the dynamic behaviour of sectoral variables. When data contain information on time for a large cross-section of sectors, traditional econometric techniques used in the macroeconomic literature such as Vector Autoregressive (VAR) and Vector Autoregressive Moving Average (VARMA) models are not appropriate since they require the estimation of too many parameters. This is why new methods which allow for the reduction of the parameter space need to be developed. The objective of this paper is both methodological and descriptive. At the methodological level we develop a simple framework for the dynamic analysis of large cross-sections. The basic model is a dynamic factor analytic model as in Sargent and Sims (1977). The sectoral variables are decomposed into two unobservable components: a common component, driven by macroeconomic shocks, and a purely sectoral component. When

Mario Forni

Papers

The Generalized Dynamic-Factor Model: Identification and Estimation

The Generalized Dynamic Factor Model one-sided estimation and forecasting

The Generalized Dynamic Factor Model: One-Sided Estimation and Forecasting

The generalized dynamic factor model: representation theory

Let's Get Real: A Factor Analytical Approach to Disaggregated Business Cycle Dynamics