Showing papers in "Journal of Applied Econometrics in 1989"

The dynamics of exchange rate volatility: A multivariate latent factor ARCH model

Abstract: We study temporal volatility patterns in seven nominal dollar spot exchange rates, all of which display strong evidence of autoregressive conditional heteroskedasticity (ARCH). We first formulate and estimate univariate models, the results of which are subsequently used to guide specification of a multivariate model. The key element of our multivariate approach is exploitation of factor structure, which facilitates tractable estimation via a substantial reduction in the number of parameters to be estimated. Such a latent-variable model is shown to provide a good description of multivariate exchange rate movements: the ARCH effects capture volatility clustering, and the factor structure captures commonality in volatility movements across exchange rates.

Investigation of production, sales and inventory relationships using multicointegration and non-symmetric error correction models

Abstract: The accumulated sum of a stationary series is called integrated. If a linear combination of some of the integrated series is stationary, it is said to be cointegrated. This paper presents empirical results on inventories to consider the possibility of a deeper form of cointegration called multicointegration, which is introduced in Granger and Lee (1989). A vector integrated series is multicointegrated if the accumulated sum of its stationary (cointegrated) linear combination is again cointegrated with itself. Inventory, which is the accumulated sum of production minus sales, is probably cointegrated with production and sales. The empirical results generally support the presence of multicointegration of production and sales in many U.S. industries and industrial aggregates. The results also favour the non-symmetric error correction model, providing evidence that the strength of attraction is different on both sides of the attractor. A modified (S, s) rule for inventory control is investigated in the context of a non-symmetric error correction model, and the results generally do not support the (S, s) rule. Sufficient evidence is found to conclude that multicointegration is a useful concept in the area of inventory determination.

A double‐hurdle model of cigarette consumption

Abstract: This paper shows the importance of the double-hurdle approach for modelling individuals' cigarette consumption, using data from the UK General Household Survey, and argues that participation and consumption should be treated as separate individual choices. The likelihood function for the full double-hurdle is derived, and it is shown how restrictions on the stochastic specification of the model and auxillary information, which identifies ex-smokers, allow it to be decomposed. The empirical results highlight the value of the sample separation information and the need to model starting and quitting as separate decisions.

Diagnostic tests for models based on individual data: A survey

Abstract: This paper surveys the growing literature on diagnostic testing of models based on unit record data. We argue that while many of these tests are produced in a Lagrange multiplier framework they are often more readily derived, and more easily applied, if approached from the conditional moment testing view of Newey (1985) and Tauchen (1985). In addition we propose some new tests based on comparisons of parametric estimators with nonparametric estimators which are consistent under certain forms of misspecification. To illustrate the utility of the tests we employ them in the examination of some existing published studies.

Output and unemployment dynamics in the United States: 1950–1985

Abstract: For US data over 1950–1985 the stochastic components of GNP growth and the unemployment rate appear to be stationary, and there is substantial feedback between these variables. The unconditional mean rate of unemployment in a joint model thus provides a natural benchmark in discussions of the ‘business cycle’. A bivariate VAR model is used to describe output–unemployment dynamics, to estimate the degree of persistence of output innovations, and to decompose output into trend and cycle. The bivariate results are interpreted using a restricted VAR and it is shown that a closely related cyclical measure can be obtained directly from the Okun's Law equation.

Strategic bargaining models and interpretation of strike data

Abstract: Attrition, screening, and signalling models of strategic bargaining are characterized in terms of their predictions about the incidence, mean duration, and settlement rates of strikes and the terms of wage settlements. These predictions are compared with the general features observed in empirical studies of strikes in Canada and the United States. Conclusions are drawn about the types of models capable of generating these features, and about the conformity of the models to the evidence. Methods are described for computing the numerical examples used to illustrate the models.

Tests of international CAPM with time-varying covariances

Abstract: We perform maximum-likelihood estimation of a model of international asset pricing based on CAPM. We test the restrictions imposed by CAPM against a more general asset pricing model. The ‘betas’ in our CAPM vary over time as the supplies of assets change and as the conditional covariances or returns on those assets change. We let the covariances change over time as a function of macroeconomic data, and an alternative model allows the covariances to follow a multivariate ARCH process. We also can identify a modified CAPM model with measurement error. We find that the estimated CAPM performs much better when variances are not constant over time. Nonetheless, CAPM is rejected in favour of the lessrestricted model of asset pricing.

Models of military expenditure

Abstract: A brief review of the extensive empirical literature devoted to explaining levels of military expenditure indicates a wide variety of unreconciled results. However, comparing the alternative models is not straightforward. This paper uses British post-war military expenditures to illustrate some of the methodological issues involved in model comparison, evaluation and selection. Starting from two published models for the UK, a systematic specification search leads to a new model, which dominates alternative specifications, passes a wide range of misspecification tests, and also fits well to French data.

The climacteric in late victorian Britain and France: A reappraisal of the evidence

Abstract: The paper investigates the question of whether the trend rate of growth of GDP slowed down in Britain or France in the nineteenth century. We establish that real GDP belongs to the trend-stationary process class of time-series in each case rather than the difference-stationary process and using Kalman filter techniques, analyse trends in a model incorporating time-varying parameters without specifying breakpoints in advance. We find, contrary to the literature, that neither country experienced a climacteric. Our results suggest that economic historians have been prone to exaggerated views of variations in trend economic growth.

Fiscal and monetary policy co-ordination: A differential game approach

Abstract: In this paper the interaction between the Treasury and the central bank is examined in the case of both cooperative and non-cooperative behaviour. Differential games are used in the framework of a continuous-time econometric model of the Italian economy. The Nash and the Stackelberg non-cooperative equilibrium solutions are computed, and the case for cooperation is analysed by considering the Nash and the Kalai-Smorodinsky bargaining models. It is shown that, in the Italian case, the government has a stronger bargaining position than the central bank. A comparison is then made between the different solutions to show that the drawbacks that emerge from non-cooperation are not simply those depending on the players' payoffs. Other features are in fact considered which constitute a further argument for policy co-ordination.

A Monte Carlo comparison of semiparametric Tobit estimators

Abstract: This paper focuses on a performance comparison of semiparametric Tobit estimators. Firstly, a conditional expectation version of Horowitz's distribution-free least-squares estimator is proposed, together with a short description of the other estimators considered in the later Monte Carlo experiment. Then, a performance comparison of the following selected estimators is made through a Monte Carlo experiment: the standard Tobit maximum-likelihood estimator, the Buckley–James estimator, Horowitz's distribution-free least-squares estimator, a conditional version of Horowitz's estimator and Powell's least absolute deviations estimator. An empirical example of Engel curve estimation with zero expenditures follows.

Does school matter for educational achievement? a two-way nested-error components analysis

Abstract: School effectiveness is a much-debated question both at an analytical level and as a policy issue. In this paper we used a two-way nested-error components educational production function model to evaluate the relative importance of latent class and school variables in explaining student school achievement. The analysis is carried out with a sample of Montreal francophone public elementary school students. As for the observable variables, the empirical results suggest that the latent class and school variables are negligible with respect to the latent personal and socioeconomic variables. Related policy issues are discussed in the conclusion.

Specification testing in censored regression models: Parametric and semiparametric methods

Abstract: We define a hierarchy of parametric and semiparametric specifications for censored regression models that is ordered according to the strength of the assumptions that are made. We review the estimation techniques and specification tests that are available at each level of the hierarchy. The estimation and testing procedures are illustrated by applying them to a model of strike duration. Several tests (some graphical) are able to detect errors in standard parametric specifications even in the presence of fairly heavy censoring, but two distinct semiparametric specifications leading to different substantive inferences cannot be rejected. In addition, the conditional mean as a prediction of the dependent variable is highly sensitive to specification errors, whereas the conditional median is not.

Production of public safety: Are socioeconomic characteristics of local communities important factors?

Abstract: This paper uses the translog cost function and cross section data from Florida to investigate the importance of socioeconomic characteristics (SEC) of local communities in the production of public safety. The approach used is noteworthy in four respects: (1) the underlying production function is specified as an AGEM production function; (2) SEC are treated as non-purchased fixed inputs; (3) two variables are used to proxy SEC in estimation; and (4) the paper test for the existence of endogenous quality differences in safety. It is shown that exclusion of SEC from the production (cost) function of safety leads to misspecification; and hence they provide a misleading guide to policy.

Self-Selection in the Residential Electricity Time-of-Use Pricing Experiments: Special Issue on Topics in Applied Econometrics

Abstract: In none of the existing studies that attempt to pool the results of the five best residential electricity TOU-pricing experiments is the problem of selection bias addressed. Three of the five experiments used samples of volunteers and it is well known that this can bias results if the ultimate inference to be made applies to the population of non-volunteers (i.e. the situation of mandatory TOU-pricing). In this paper we adapt some well-used methodology for correcting for self-selection to the five experiments in question and estimate the extent to which its presence biases estimates of the elasticity of substitution between peak and off-peak electricity consumption. We find, for example, that the bias is upwards of + 24 per cent in the Los Angeles experiment, resulting in a substantial overstatement of the response of customers to mandatory TOU prices. While we stop short of integrating our methodology into a full-blown pooling framework, users of the EPRI-sponsored RETOU program should nevertheless be cautioned to the potential biases inherent in forecasts of TOU response derived from it, since no account for the selfselection phenomenon and its consequences was taken in its development.

Self-selection in the residential electricity time-of-use pricing experiments

Abstract: In none of the existing studies that attempt to pool the results of the five best residential electricity TOU-pricing experiments is the problem of selection bias addressed. Three of the five experiments used samples of volunteers and it is well known that this can bias results if the ultimate inference to be made applies to the population of non-volunteers (i.e. the situation of mandatory TOU-pricing). In this paper we adapt some well-used methodology for correcting for self-selection to the five experiments in question and estimate the extent to which its presence biases estimates of the elasticity of substitution between peak and off-peak electricity consumption. We find, for example, that the bias is upwards of + 24 per cent in the Los Angeles experiment, resulting in a substantial overstatement of the response of customers to mandatory TOU prices. While we stop short of integrating our methodology into a full-blown pooling framework, users of the EPRI-sponsored RETOU program should nevertheless be cautioned to the potential biases inherent in forecasts of TOU response derived from it, since no account for the selfselection phenomenon and its consequences was taken in its development.

Monetary policy and the stability of macroeconomic relationships

Abstract: SUMMARY Estimates of the effect of different international monetary regimes on the parameters of the Phillips curve, the Keynesian consumption function, and other reduced-form macroeconomic relationships are given. The estimates provide a quantitative assessment of the importance of the Lucas critique for such regime shifts. The estimates are calculated by stochastically simulating an estimated multicountry economic model with rational expectations under a fixed exchange rate regime and a flexible exchange rate regime. In both regimes interest rates are the primary instrument of monetary policy. Noticeable shifts occur in most of the macroeconomic relationships, especially in the consumption function and the Phillips curve, and these shifts have simple economic interpretations based on the changes in the variance and the serial correlation of income and prices in the two regimes. However, in most cases the shifts are not large quantitatively. At least, as implemented here, this type of regime shift does not seem to generate much instability in the conventional macroeconomic relationships. Several possible reasons for this finding are discussed in the paper. In this paper I plan to follow in a tradition pioneered by A. W. Phillips almost 40 years ago. Phillips (1954) evaluated monetary and fiscal policy by examining the operating properties of simple policy rules in a fully specified economic model. He did this by applying control theory ideas from the engineering literature. By examining the performance of the model economy under different types of policy rules-for example, proportional, derivative, and integral rules-he attempted to determine which policy rule would work well and which would not work well in practice. Following in the Phillips tradition does not mean that I will use the same theoretical models, econometric methods, or computational techniques that Phillips used. To do so would, in my view, pass by fundamental developments in dynamic economic theory, econometrics, and solution procedures that have been achieved since the time that Phillips wrote. Phillip's (1954) analysis was based on Keynesian ISLM-multiplier-accelerator models that were popular in the early 1950s. The policy evaluation I describe in this lecture differs from Phillips's policy evaluation in several ways. First, it makes use of a dynamic rational expectations model with forward-looking behaviour in consumption, investment, wage-setting, and interest rates. Second, the model is international and exchange rate determination reflects the current high degree of international capital mobility which did not exist when Phillips wrote. Third, the model is empirically estimated; most of Phillips's work was based on simulation models with 'calibrated' parameters. Fourth, the simulations are done stochastically with an empirically

A random walk through the Gibson paradox

Abstract: Evidence to support the Gibson paradox is often given in the form of a simple correlation between the nominal interest rate and the log of price level, or in the form of a simple linear regression between these two variables. Authors then show, using standard procedures of statistical inference, that the price level possesses a significant coefficient. We argue that this class of evidence is spurious since the nominal interest rate and the price level (both integrated variables) do not form a cointegrated system.

A qualitative response analysis of UK firms' employment and output decisions

Abstract: This paper investigates firms' employment and output decisions and presents some empirical evidence concerning the rationality of firms' expectations. The dymanic model is based on the assumptions of convex adjustment costs and monopolistic competition in the product market. The results are obtained using categorical information on individual firms contained in business surveys collected by the Confederation of British Industry. Ordered probit models are used to estimate the employment and output equations. We find that expected demand is a critical determinant of firms' decisions, the effect of changes in cost conditions is not as well determined, and the data are not consistent with the rational expectations hypothesis.

Ex‐post rational price approximations and the empirical reliability of the present‐value relation

Abstract: Excess volatility and regression tests have resulted in apparent rejections of the present-value relation when ex-post price approximations are employed. These approximations are based upon a sample terminal condition for prices, are not ergodic time-series, and do not result in statistics with readily calculable standard errors. Kleidon (1986a) has demonstrated that ex-post price approximations can subtly affect the reliability of certain volatility tests. We use a bootstrapped cointegration model to demonstrate some of these same effects in Mankiw, Romer and Shapiro's (1985) volatility statistics. The volatility statistics rarely have positive expected value in finite samples and still do not reject the presentvalue relation. Approximations based upon a ‘rolling’ terminal condition result in volatility statistics which have calculable large-sample errors, but even these standard errors greatly overstate the accuracy of volatility statisics in small samples. Regression tests of the present value relation are also affected by the price approximations.

Differences in the properties of large‐scale macroeconometric models: The role of labour market specifications

Abstract: This paper examines important differences in simulation properties of four leading UK macroeconometric models that arise from features of their labour market specifications. Econometric evaluation of model equations and their subsequent re-specification, together with sensitivity analysis of simulation results, explains and resolves differences in overall model properties. The examples cover the determination of non-manufacturing employment, the modelling of employment, unemployment and the working population, and the effect of the retention ratio on wage determination. Increased attention to systematic comparison and testing leads to increased convergence of model simulation results or, as second best, increased understanding of the reasons for divergence.

Survival analysis software on MS/PC‐DOS computers

Abstract: There exist more than a dozen MS/PC-DOS programs with the capability of performing some form of survival analysis. Most of these are reviewed here, with respect to (1) the survival analysis methods they cover, (2) their ease of use and flexibility, (3) their user interface, (4) their graphics capabilities, and (5) their computational accuracy.

Constructing the true cost of living index from the Engel curves of the PIGLOG model

Abstract: This paper investigates the properties of the Tornqvist index and shows how this can be used to construct a true cost of living index from a simple system of Engel curves. This allows the substitution bias inherent in a fixed weight index to be estimated using the maximum degree of commodity disaggregation available in the data. Furthermore the proposed index can be decomposed to show the impact of inflation on households with different expenditure levels and household chararacteristics. An empirical application of the method is given, using data on UK food expenditures.

An examination of the equilibrium specification and structure of production for Canadian telecommunications

Abstract: Multiple-output models of Canadian telecommunications production are estimated under different production equilibria. A specification test is conducted between the short- and long-run equilibrium models and the long-run equilibrium is rejected. In order to capture the nature of the disequilibrium, a dynamic cost of adjustment model is estimated for Bell Canada. There are significant adjustment costs and it is estimated that for $1.00 of marginal capital costs the carrier must incur an additional cost of $0.30 to install the new capital into the production process.

Structural time series analysis and modelling package: A review

Abstract: This package performs specialized tasks related to specification, estimation, prediction and diagnostic checking in the context of a particular class of unobserved-components models. It is menu-driven and easy to use. Although the present (first) release suffers from a number of limitations, omissions and bugs, it is nevertheless a useful tool in the hands of a knowledgeable user.

The relationship between the level and variability of inflation: Theory and evidence

Abstract: SUMMARY This paper emphasizes the role played by asymmetric stabilization policies in producing the widely reported positive association between the level and variability of inflation. A simple theoretical model is constructed which shows how asymmetric policies give rise to this association. The empirical findings presented provide support to the proposition that variations in inflation uncertainty are policy-induced.

Statistical properties of data stretching

Abstract: This research examines the properties of an estimation procedure frequently used because observations on some variables are available only at higher levels of aggregation than others. When this occurs, data are often stretched by repeating observations on variables at higher levels of aggregation. We show that this procedure results in biased estimators of coefficients and error variances. Under some circumstances the estimation based on stretched data has a smaller covariance matrix than that based on aggregated data. Comparisons of mean squared errors depend on unknown coefficients.

Econometric modelling on microcomputers: A review of major software packages

Abstract: In their present state, both in terms of price and power, microcomputers are taking over applied econometics. Now even model simulation is feasible. This paper tries to help the model-builder in determining which of the following modelling software is appropriate to his/her specific needs: AREMOS, ESP, Micro-TSP, MODLER and SORITEC. Tables devoted to data management, estimation, model specification, simulation, and presentation of results, detail the availability of desirable features. Finally, individual short reviews stress the best and worst points of each software.

Planning, criticism, and revision

Abstract: This paper presents a more complete theory of data analysis which allows for changes in the state of mind of the observer and also for approximations that limit planning costs. Discussion is included on the form that criticism should take, and the extent to which planned responses to the data can legitimately be revised after the data are reviewed. The proper role of diagnostics is discussed. Some diagnostic statistics are genuinely criticisms, but many are pre-test diagnostics that play a role in a complex multi-step method of estimation. A third category is elicitation diagnostics, which ask data-dependent questions about the prior distribution.