
on Econometrics 
By:  Yukitoshi Matsushita (Graduate School of Economics, University of Tokyo) 
Abstract:  This paper studies the properties of Likelihood Ratio (LR) tests associated with the limited information maximum likelihood (LIML) estimators in a structural form estimation when the number of instrumental variables is large. Two types of asymptotic theories are developed to approximate the distribution of the likelihood ratio (LR) statistics under the null hypothesis H0 : ƒÀ = ƒÀ0: the (large sample) asymptotic expansion and the largeKn asymptotic theory. The size comparison of two modified LR tests based on these two asymptotics is made with Moreira's conditional likelihood ratio (CLR) test and the large K ttest. 
Date:  2007–02 
URL:  http://d.repec.org/n?u=RePEc:tky:fseres:2007cf466&r=ecm 
By:  Yukitoshi Matsushita (Graduate School of Economics, University of Tokyo) 
Abstract:  This paper studies the properties of tratios associated with the limited information maximum likelihood (LIML) estimators in a structural form estimation when the number of instrumental variables is large. Asymptotic expansions are made of the distributions of a large K tratio statistic under largeKn asymptotics. A modified tratio statistic is proposed from the asymptotic expansion. The power of the large K tratio test dominates the AR test, the Ktest by Kleibergen (2002), and the conditional LR test by Moreira (2003); and the difference can be substantial when the instruments are weak. 
Date:  2007–02 
URL:  http://d.repec.org/n?u=RePEc:tky:fseres:2007cf467&r=ecm 
By:  Feng, Yuanhua; Yu, Keming 
Abstract:  A new multivariate random walk model with slowly changing parameters is introduced and investigated in detail. Nonparametric estimation of local covariance matrix is proposed. The asymptotic distributions, including asymptotic biases, variances and covariances of the proposed estimators are obtained. The properties of the estimated value of a weighted sum of individual nonparametric estimators are also studied in detail. The integrated effect of the estimation errors from the estimation for the difference series to the integrated processes is derived. Practical relevance of the model and estimation is illustrated by application to several foreign exchange rates. 
Keywords:  Multivariate time series; slowly changing vector random walk; local covariance matrix; kernel estimation; asymptotic properties; forecasting. 
JEL:  C32 G00 C14 
Date:  2006 
URL:  http://d.repec.org/n?u=RePEc:pra:mprapa:1597&r=ecm 
By:  Masato Ubukata (Graduate School of Economics, Osaka University); Kosuke Oya (Graduate School of Economics, Osaka University) 
Abstract:  The cumulative covariance estimator in Hayashi and Yoshida (2005b) which suits for nonsynchronous observations possibly has a bias in the presence of the observational noise. We propose the test statistic to detect whether the observational noise causes a measurable bias in the estimator of Hayashi and Yoshida (2005b). The test statistic proposed in this paper is asymptotically distributed as standard normal under null hypothesis. The finite sample performance of the test statistic is investigated through Monte Carlo simulation. 
Keywords:  test statistic; integrated covariance; nonsynchronous observation; observational noise; market microstructure noise 
JEL:  C12 D49 
Date:  2007–02 
URL:  http://d.repec.org/n?u=RePEc:osk:wpaper:0703&r=ecm 
By:  Silvennoinen, Annastiina (School of Finance and Economics); Teräsvirta, Timo (School of Economics and Management) 
Abstract:  In this paper we propose a multivariate GARCH model with a timevarying conditional correlation structure. The new Double Smooth Transition Conditional Correlation GARCH model extends the Smooth Transition Conditional Correlation GARCH model of Silvennoinen and Teräsvirta (2005) by including another variable according to which the correlations change smoothly between states of constant correlations. A Lagrange multiplier test is derived to test the constancy of correlations against the DSTCCGARCH model, and another one to test for another transition in the STCCGARCH framework. In addition, other specification tests, with the aim of aiding the model building procedure, are considered. Analytical expressions for the test statistics and the required derivatives are provided. The model is applied to a selection of world stock indices, and it is found that time is an important factor affecting correlations between them. 
Keywords:  Multivariate GARCH; Constant conditional correlation; Dynamic conditional correlation; Return comovement; Variable correlation GARCH model; Volatility model evaluation 
JEL:  C12 C32 C51 C52 G10 
Date:  2007–02–01 
URL:  http://d.repec.org/n?u=RePEc:hhs:hastef:0652&r=ecm 
By:  Alkhamisi, M.A. (Department of Mathematics, Salahaddin University, KurdistanRegion, Iraq); Shukur, Ghazi (Departments of Economics and Statistics, Jönköping International Business School (JIBS), Sweden) 
Abstract:  In this paper, a number of procedures have been proposed for developing new biased estimators of seemingly unrelated regression (SUR) parameters, when the explanatory variables are affected by multicollinearity. Several ridge parameters are proposed and then compared in terms of the trace mean squared error (TMSE) and(PR) criterion. The PR is the proportion of replication (out of 1,000) for which the SUR version of the generalised least squares, (SGLS) estimator has a smaller TMSE than the others. The study has been made using Monte Carlo simulations where the number of equations in the system, number of observations, correlation among equations and correlation between explanatory variables have been varied. For each model we performed 1,000 replications. Our results show that under certain conditions the performance of the multivariate regression estimators based on SUR ridge parameters RSarith, RSqarith and RSmax are superior to other estimators in terms of TMSE and PR criterion. In large samples and when the collinearity between the explanatory variables is not high the unbiased SUR, estimator produces a smaller TMSEs. 
Keywords:  Multicollinearity; SUR ridge regression; Monte Carlo simulations; biased estimators; Generalized least squares 
JEL:  C30 C51 C52 
Date:  2007–01–31 
URL:  http://d.repec.org/n?u=RePEc:hhs:cesisp:0080&r=ecm 
By:  John W. Galbraith; Greg Tkacz 
Abstract:  For stationary transformations of variables, there exists a maximum horizon beyond which forecasts can provide no more information about the variable than is present in the unconditional mean. Meteorological forecasts, typically excepting only experimental or exploratory situations, are not reported beyond this horizon; by contrast, little generally accepted information about such maximum horizons is available for economic variables. The authors estimate such content horizons for a variety of economic variables, and compare these with the maximum horizons that they observe reported in a large sample of empirical economic forecasting studies. The authors find that many published studies provide forecasts exceeding, often by substantial margins, their estimates of the content horizon for the particular variable and frequency. The authors suggest some simple reporting practices for forecasts that could potentially bring greater transparency to the process of making and interpreting economic forecasts. 
Keywords:  Econometric and statistical methods, Business fluctuations and cycles 
JEL:  C53 
Date:  2007 
URL:  http://d.repec.org/n?u=RePEc:bca:bocawp:071&r=ecm 
By:  Feng, Yuanhua; Beran, Jan; Yu, Keming 
Abstract:  A class of semiparametric fractional autoregressive GARCH models (SEMIFARGARCH), which includes deterministic trends, difference stationarity and stationarity with short and longrange dependence, and heteroskedastic model errors, is very powerful for modelling financial time series. This paper discusses the model fitting, including an efficient algorithm and parameter estimation of GARCH error term. So that the model can be applied in practice. We then illustrate the model and estimation methods with a few of different finance data sets. 
Keywords:  Financial time series; GARCH model; SEMIFAR model; parameter estimation; kernel estimation; asymptotic property. 
JEL:  G00 C22 C14 
Date:  2006 
URL:  http://d.repec.org/n?u=RePEc:pra:mprapa:1593&r=ecm 
By:  Susumu Imai (Queen's University); Neelam Jain (Northern Illinois University); Andrew Ching (University of Toronto) 
Abstract:  We propose a new methodology for structural estimation of dynamic discrete choice models. We combine the Dynamic Programming (DP) solution algorithm with the Bayesian Markov Chain Monte Carlo algorithm into a single algorithm that solves the DP problem and estimates the parameters simultaneously. As a result, the computational burden of estimating a dynamic model becomes comparable to that of a static model. Another feature of our algorithm is that even though per solutionestimation iteration, the number of grid points on the state variable is small, the number of effective grid points increases with the number of estimation iterations. This is how we help ease the "Curse of Dimensionality". We simulate and estimate several versions of a simple model of entry and exit to illustrate our methodology. We also prove that under standard conditions, the parameters converge in probability to the true posterior distribution, regardless of the starting values. 
Keywords:  Bayesian Estimation, Dynamic Discrete Choice Model, Dynamic Programming, Markov Chain Monte Carlo, Bayesian Dynamic Programming Estimation 
JEL:  C51 C61 C63 L00 
Date:  2006–12 
URL:  http://d.repec.org/n?u=RePEc:qed:wpaper:1118&r=ecm 
By:  Minford, Patrick (Cardiff Business School); Theodoridis, Konstantinos (Cardiff Business School); Meenagh, David (Cardiff Business School) 
Abstract:  In this article we introduce a new bootstrap method for testing DSGE models according to their dynamic performance. The method maintains a separation between the structural (nonlinear) model as the null hypothesis and its dynamic time series representation. The model's errors are discovered and used for bootstrapping (after whitening); the resulting pseudosamples are used to discover the sampling distribution of the dynamic time series model. The test then establishes whether the parameters of the timeseries model estimated on the actual data lie within some confidence interval of this distribution. A Waldtype statistic is developed for this purpose. 
Date:  2007–01 
URL:  http://d.repec.org/n?u=RePEc:cdf:wpaper:2007/2&r=ecm 
By:  Feng, Yuanhua 
Abstract:  This paper introduces the idea that the variances or correlations in financial returns may all change conditionally and slowly over time. A multistep local dynamic conditional correlation model is proposed for simultaneously modelling these components. In particular, the local and conditional correlations are jointly estimated by multivariate kernel regression. A multivariate kNN method with variable bandwidths is developed to solve the curse of dimension problem. Asymptotic properties of the estimators are discussed in detail. Practical performance of the model is illustrated by applications to foreign exchange rates. 
Keywords:  Local and conditional correlations; multivariate nonparametric ARCH; multivariate kernel regression; multivariate kNN method. 
JEL:  G0 G1 C32 
Date:  2006 
URL:  http://d.repec.org/n?u=RePEc:pra:mprapa:1592&r=ecm 
By:  Juan Carlos Escanciano (Department of Economics, Indiana University); Kyungchul Song (Department of Economics, University of Pennsylvania) 
Abstract:  This paper proposes an asymptotically optimal specification test of singleindex models against alternatives that lead to inconsistent estimates of a covariate’s average partial effect. The proposed tests are relevant when a researcher is concerned about a potential violation of the singleindex restriction only to the extent that the estimated average partial effects suffer from a nontrivial bias due to the misspecifcation. Using a pseudonorm of average partial effects deviation and drawing on the minimax approach, we find a nice characterization of the least favorable local alternatives associated with misspecified average partial effects as a single direction of Pitman local alternatives. Based on this characterization, we define an asymptotic optimal test to be a semiparametrically efficient test that tests the significance of the least favorable direction in an augmented regression formulation, and propose such a one that is asymptotically distributionfree, with asymptotic critical values available from the X 2/1 table. The testing procedure can be easily modified when one wants to consider average partial effects with respect to binary covariates or multivariate average partial effects. 
Keywords:  Average Partial Effects, Omnibus tests, Optimal tests, Semi parametric Efficiency, Efficient Score 
JEL:  C14 
Date:  2007–01–29 
URL:  http://d.repec.org/n?u=RePEc:pen:papers:07005&r=ecm 
By:  Roy Surupa; Banerjee Tathagata 
Abstract:  The focus of this paper is on regression models for mixed binary and continuous outcomes, when the true predictor is measured with error and the binary responses are subject to classification errors. Latent variable is used to model the binary response. The joint distribution is expressed as a product of the marginal distribution of the continuous response and the conditional distribution of the binary response given the continuous response. Models are proposed to incorporate the measurement error and/or classification errors. Likelihood based analysis is performed to estimate the regression parameters of interest. Theoretical studies are made to find the bias of the likelihood estimates of the model parameters. An extensive simulation study is carried out to investigate the effect of ignoring classification errors and/or measurement error on the estimates of the model parameters. The methodology is illustrated with a data set obtained by conducting a small scale survey. 
Keywords:  mixed binarycontinuous outcomes; classification errors; Berkson model; Maximum likelihood estimate; Misspecified model 
Date:  2007–01–25 
URL:  http://d.repec.org/n?u=RePEc:iim:iimawp:20070108&r=ecm 
By:  Juan Carlos Escanciano (Indiana University); Silvia Mayoral (Universidad de Navarra) 
Abstract:  A general method for testing the martingale difference hypothesis is proposed. The new tests are datadriven smooth tests based on the principal components of certain marked empirical processes that are asymptotically distributionfree, with critical values that are already tabulated. The datadriven smooth tests are optimal in a semiparametric sense discussed in the paper, and they are robust to conditional heteroskedasticity of unknown form. A simulation study shows that the smooth tests perform very well for a wide range of realistic alternatives and have more power than the omnibus and other competing tests. Finally, an application to the S&P 500 stock index and some of its components highlights the merits of our approach. 
URL:  http://d.repec.org/n?u=RePEc:una:unccee:wp0107&r=ecm 
By:  John M Maheu; Thomas H McCurdy 
Abstract:  We propose a new discretetime model of returns in which jumps capture persistence in the conditional variance and higherorder moments. Jump arrival is governed by a heterogeneous Poisson process. The intensity is directed by a latent stochastic autoregressive process, while the jumpsize distribution allows for conditional heteroskedasticity. Model evaluation focuses on the dynamics of the conditional distribution of returns using density and variance forecasts. Predictive likelihoods provide a periodbyperiod comparison of the performance of our heterogeneous jump model relative to conventional SV and GARCH models. Further, in contrast to previous studies on the importance of jumps, we utilize realized volatility to assess outofsample variance forecasts. 
Keywords:  jump clustering, jump dynamics, MCMC, predictive likelihood, realized volatility, Bayesian model average 
JEL:  C22 C11 G1 
Date:  2007–02–02 
URL:  http://d.repec.org/n?u=RePEc:tor:tecipa:tecipa279&r=ecm 
By:  João Victor Issler (EPGE/FGV); Luiz Renato Regis de Oliveira Lima (EPGE/FGV) 
Date:  2007–01 
URL:  http://d.repec.org/n?u=RePEc:fgv:epgewp:642&r=ecm 
By:  Martellosio, Federico 
Abstract:  Many popular tests for residual spatial autocorrelation in the context of the linear regression model belong to the class of invariant tests. This paper derives some exact properties of the power function of such tests. In particular, we characterize the circumstances under which the limiting power, as the autocorrelation increases, vanishes, thus extending the work of Krämer (2005, Journal of Statistical Planning and Inference 128, 489496). More generally, the analysis in the paper sheds new light on how the power of invariant tests for spatial autocorrelation is affected by the matrix of regressors and by the spatial structure. A numerical study aimed at assessing the practical relevance of the theoretical results is included. 
Keywords:  CliffOrd test; invariant tests; linear regression model; point optimal tests; power; similar tests; spatial autocorrelation 
JEL:  C21 C12 
Date:  2006–12 
URL:  http://d.repec.org/n?u=RePEc:pra:mprapa:1697&r=ecm 
By:  Tatsuya Kubokawa (Faculty of Economics, University of Tokyo) 
Abstract:  Sample survey data can be used to derive a reliable estimate of a total mean for a large area. When the same data are used to estimate means of small areas like city, county or town belonging to the large area, the usual direct estimators like the sample mean have unacceptably large standard errors due to the small sizes of the samples in the small areas. This is called a small area problem. To find more accurate estimates for given small areas, one needs to "borrow strength" from the related areas. The linear mixed model (LMM) is recognized as an appropriate model for handling such a problem, and the resulting empirical best linear unbiased predictor (EBLUP) can yield a smaller standard error. This article gives a review of the small area estimation based on LMM. Especially, the article explains how the structure of (common parameters)+(random effects) in LMM works to get accurate estimates. The estimators of the mean squared errors of EBLUP and the confidence interval based on EBLUP are derived to evaluate accuracy of EBLUP. Finally, some generalizations and various variants of LMM are described for analyzing spatial data, and the generalized linear mixed model (GLMM) and its application to estimation of mortality rates are explained. 
Date:  2006–12 
URL:  http://d.repec.org/n?u=RePEc:tky:jseres:2006cj171&r=ecm 
By:  Olivier Parent; James P. Lesage 
Abstract:  We extend the literature on Bayesian model comparison for ordinary leastsquares regression models to include spatial autoregressive and spatial error models. Our focus is on comparing models that consist of different matrices of explanatory variables. A Markov Chain Monte Carlo model composition methodology labelled MC to the third by Madigan and York (1995) is developed for two types of spatial econometric models that are frequently used in the literature. The methodology deals with cases where the number of possible models based on different combinations of candidate explanatory variables is large enough that calculation of posterior probabilities for all models is difficult or infeasible. Estimates and inferences are produced by averaging over models using the posterior model probabilities as weights, a procedure known as Bayesian model averaging. We illustrate the methods using a spatial econometric model of origindestination population migration flows between the 48 US States and District of Columbia during the 1990 to 2000 period. 
Date:  2007 
URL:  http://d.repec.org/n?u=RePEc:cin:ucecwp:200702&r=ecm 
By:  Jürgen Franke; JeanPierre Stockis; Joseph Tadjuidje 
Abstract:  We consider the problem of estimating the conditional quantile of a time series at time t given observations of the same and perhaps other time series available at time t  1. We discuss sieve estimates which are a nonparametric versions of the KoenkerBassett regression quantiles and do not require the specification of the innovation law. We prove consistency of those estimates and illustrate their good performance for light and heavytailed distributions of the innovations with a small simulation study. As an economic application, we use the estimates for calculating the value at risk of some stock price series. 
Keywords:  Conditional Quantile, Time Series, Sieve Estimate, Neural Network, Qualitative Threshold Model, Uniform Consistency, Value at Risk 
JEL:  C14 C45 
Date:  2007–02 
URL:  http://d.repec.org/n?u=RePEc:hum:wpaper:sfb649dp2007005&r=ecm 
By:  Tomas del Barrio Castro (Universitat de Barcelona) 
Keywords:  hegy tests, vector of quarters, unit root tests, seasonality 
JEL:  C22 C12 
Date:  2007 
URL:  http://d.repec.org/n?u=RePEc:bar:bedcje:2007170&r=ecm 
By:  A. PRINZIE; D. VAN DEN POEL 
Abstract:  Several supervised learning algorithms are suited to classify instances into a multiclass value space. MultiNomial Logit (MNL) is recognized as a robust classifier and is commonly applied within the CRM (Customer Relationship Management) domain. Unfortunately, to date, it is unable to handle huge feature spaces typical of CRM applications. Hence, the analyst is forced to immerse himself into feature selection. Surprisingly, in sharp contrast with binary logit, current software packages lack any feature selection algorithm for MultiNomial Logit. Conversely, Random Forests, another algorithm learning multi class problems, is just like MNL robust but unlike MNL it easily handles highdimensional feature spaces. This paper investigates the potential of applying the Random Forests principles to the MNL framework. We propose the Random MultiNomial Logit (RMNL), i.e. a random forest of MNLs, and compare its predictive performance to that of a) MNL with expert feature selection, b) Random Forests of classification trees. We illustrate the Random MultiNomial Logit on a crosssell CRM problem within the homeappliances industry. The results indicate a substantial increase in model accuracy of the RMNL model to that of the MNL model with expert feature selection. 
Keywords:  multiclass classifier design and evaluation, feature evaluation and selection, data mining methods and algorithms, customer relationship management (CRM) 
Date:  2007–01 
URL:  http://d.repec.org/n?u=RePEc:rug:rugwps:07/435&r=ecm 
By:  Kattai, Rasmus 
Abstract:  This paper focuses on the dynamic properties of error correction models (ECM). It is shown that the absence of structural breaks in the cointegrating vector does not necessarily imply that also all parameters of the dynamic specification of the ECM are time invariant. In some cases, depending on the data generating process of regressors, the intercept has to be time varying in order to have the long run equilibrium of a dynamic model independent of the growth rates of the variables out of sample period, i.e. to satisfy the dynamic homogeneity condition. It is found to be common when estimating ECMs on macroeconomic time series of converging countries. Dynamic homogeneity can be achieved by imposing the state dependent dynamic homogeneity restriction on the intercept. Applying the restriction is illustrated by an empirical example using Estonian data on real wages and labour productivity. 
Keywords:  dynamic homogeneity, error correction models, forecasting 
JEL:  C32 C51 
URL:  http://d.repec.org/n?u=RePEc:eea:boewps:wp200701&r=ecm 
By:  Martellosio, Federico 
Abstract:  This paper studies the correlation structure of spatial autoregressions defined over arbitrary configurations of observational units. We derive a number of new properties of the models and provide new interpretations of some of their known properties. A little graph theory helps to clarify how the correlation between two random variables observed at two units depends on the walks connecting the two units, and allows to discuss the statistical consequences of the presence (or, more importantly in econometrics, the absence) of symmetries or regularities in the configuration of the observational units. The analysis is centered upon oneparameter models, but extensions to multiparameter models are also considered. 
Keywords:  exponential families; graphs; quadratic subspace; spatial autoregressions; spatial weights matrices 
JEL:  C21 C50 
Date:  2006–10–16 
URL:  http://d.repec.org/n?u=RePEc:pra:mprapa:1695&r=ecm 
By:  Bruno Eklund 
Abstract:  This paper focuses on the Stock and Watson methodology to fore cast the future state of the business cycle in the Icelandic economy. By selecting variables available on a monthly basis that mimic the cyclical behaviour of the quarterly GDP, coincident and leading vari ables are identi?ed. A factor model is then speci?ed based on the assumption that a single common unobservable element drives the cyclical evolution of many of the Icelandic macroeconomic variables. The model is cast into a state space form providing a simple frame work both for estimation and for predicting the future recession and expansion patterns. Based on the bootstrap resampling technique, a simple approach to estimate recession and expansion probabilities is developed. This method is completely nonparametric compared to the semiparametric approach used by Stock and Watson. 
Date:  2007–01 
URL:  http://d.repec.org/n?u=RePEc:ice:wpaper:wp33_bruno&r=ecm 
By:  Arz, Stephanus 
Abstract:  Usually, seasonal adjustment is based on time series models which decompose an unadjusted series into the sum or the product of four unobservable components (trendcycle, seasonal, workingday and irregular components). In the case of clearly weatherdependent output in the west German construction industry, traditional considerations lead to an additive model. However, this results in an overadjustment of calendar effects. An alternative is a multiplicativeadditive mixed model, the estimation of which is illustrated using X12ARIMA. Finally, the relevance of the new model is shown by analysing selected time series for different countries. 
Keywords:  Seasonal adjustment, calendar adjustment, overadjustment, multiplicativeadditive model, X12ARIMA 
JEL:  C22 
Date:  2006 
URL:  http://d.repec.org/n?u=RePEc:zbw:bubdp1:5196&r=ecm 
By:  Fanelli, Luca 
Abstract:  When in "exact" present value (PV) relations the decision variables do not Granger cause the explanatory variables and a VAR process is used to derive restrictions, the system embodies explosive roots. Hence any test of the PV restrictions would reject the null if the system incorporates Granger noncausality constraints. This paper investigates the issue. 
Keywords:  Granger non causality; Present value model; VAR 
JEL:  C00 
Date:  2006–12 
URL:  http://d.repec.org/n?u=RePEc:pra:mprapa:1642&r=ecm 
By:  Markus Pannenberg; Martin Spiess 
Abstract:  We propose a generalized estimating equations approach to the analysis of the mean and the covariance structure of a bivariate time series process of panel data with mixed continuous and discrete dependent variables. The approach is used to jointly analyze wage dynamics and the incidence of profitsharing in West Germany. Our findings reveal a significantly positive conditional correlation of wages and the incidence of profitsharing. Furthermore, they indicate that permanent unobserved individual ability is comparatively more important in the profitsharing than in the wage equation and show that shocks have a longlasting effect on transitory wages but not on the incidence of profitsharing. Hence, the results support theoretical predictions that selection into profitsharing is mostly due to unobservable ability and that profitsharing ties wages more closely to productivity. 
Keywords:  Generalized estimating equations, covariance structure, longitudinal data, real wages, variable pay 
JEL:  C33 C35 D31 J31 J33 
Date:  2007 
URL:  http://d.repec.org/n?u=RePEc:diw:diwwpp:dp663&r=ecm 
By:  Prof John Foster (School of Economics, The University of Queensland) 
Abstract:  Applied economics has long been dominated by multiple regression techniques. In this regard, econometrics has tended to have a narrower focus than, for example, psychometrics in psychology. Over the last two decades, the simulation and calibration approach to modeling has become more popular as an alternative to traditional econometric strategies. However, in contrast to the welldeveloped methodologies that now exist in econometrics, simulation/calibration remains exploratory and provisional, both as an explanatory and as a predictive modelling technique although clear progress has recently been made in this regard (see Brenner and Werker (2006)). In this paper, we suggest an approach that can usefully integrate both of these modelling strategies into a coherent evolutionary economic methodology. 
Date:  2007 
URL:  http://d.repec.org/n?u=RePEc:qld:uq2004:343&r=ecm 
By:  Troy Matheson; James Mitchell; Brian Silverstone (Reserve Bank of New Zealand) 
Abstract:  The qualitative responses that firms give to business survey questions regarding changes in their own output provide a realtime signal of official output changes. The most commonlyused method to produce an aggregate quantitative indicator from business survey responses  the net balance, or diffusion index  has changed little in 40 years. It focuses on the proportion of survey respondents replying "up", "the same" or "down". This paper investigates whether an improved realtime signal of official output data changes can be derived from a recently advanced method on the aggregation of survey data from panel responses. It also considers the ability of survey data to anticipate revisions to official output data. We find, in a New Zealand application, that exploiting the panel dimension to qualitative survey data gives a better insample signal about official data than traditional methods. This is achieved by giving a higher weight to firms whose answers have a close link to official data than to those whose experiences correspond only weakly or not at all. Outofsample, it is less clear it matters how survey data are quantified with simpler and more parsimonious methods hard to improve. It is clear, nevertheless, that survey data, exploited in some form, help to explain revisions to official data. 
JEL:  C35 C42 C53 C80 
Date:  2007–02 
URL:  http://d.repec.org/n?u=RePEc:nzb:nzbdps:2007/02&r=ecm 