Time Series Econometric Models Research Articles

Conditional volatility forecasting in a dynamic hedging model

This paper addresses several questions surrounding volatility forecasting and its use in the estimation of optimal hedging ratios. Specifically: Are there economic gains by nesting time-series econometric models (GARCH) and dynamic programming models (therefore forecasting volatility several periods out) in the estimation of hedging ratios whilst accounting for volatility in the futures bid–ask spread? Are the forecasted hedging ratios (and wealth generated) from the nested bid–ask model statistically and economically different than standard approaches? Are there times when a trader following a basic model that does not forecast outperforms a trader using the nested bid–ask model? On all counts the results are encouraging—a trader that accounts for the bid–ask spread and forecasts volatility several periods in the nested model will incur lower transactions costs and gain significantly when the market suddenly and abruptly turns. Copyright © 2005 John Wiley & Sons, Ltd.

The impact of the us strategic defence initiative on the space race

This paper models the space race between the US and the Soviet Union during the Cold War using time series econometric modelling. The analysis shows that, post the Vietnam‐war era, military considerations play a key role in explaining the behaviour of US space expenditure. In particular, the US Department of Defense (DoD) Strategic Defense Initiative (SDI) of the early 1980s is shown to result in a changing structure of the space race. This is expected to continue in the future with the current anti‐ballistic missile (ABM) programme and the withdrawal of the US from the ABM treaty of on space militarization.

Empirical Limits for Time Series Econometric Models

This paper characterizes empirically achievable limits for time series econometric modeling and forecasting. The approach involves the concept of minimal information loss in time series regression and the paper shows how to derive bounds that delimit the proximity of empirical measures to the true probability measure (the DGP) in models that are of econometric interest. The approach utilizes joint probability measures over the combined space of parameters and observables and the results apply for models with stationary, integrated, and cointegrated data. A theorem due to Rissanen is extended so that it applies directly to probabilities about the relative likelihood (rather than averages), a new way of proving results of the Rissanen type is demonstrated, and the Rissanen theory is extended to nonstationary time series with unit roots, near unit roots, and cointegration of unknown order. The corresponding bound for the minimal information loss in empirical work is shown not to be a constant, in general, but to be proportional to the logarithm of the determinant of the (possibility stochastic) Fisher–information matrix. In fact, the bound that determines proximity to the DGP is generally path dependent, and it depends specifically on the type as well as the number of regressors. For practical purposes, the proximity bound has the asymptotic form (K/2)log n, where K is a new dimensionality factor that depends on the nature of the data as well as the number of parameters in the model. When ‘good’ model selection principles are employed in modeling time series data, we are able to show that our proximity bound quantifies empirical limits even in situations where the models may be incorrectly specified. One of the main implications of the new result is that time trends are more costly than stochastic trends, which are more costly in turn than stationary regressors in achieving proximity to the true density. Thus, in a very real sense and quantifiable manner, the DGP is more elusive when there is nonstationarity in the data. The implications for prediction are explored and a second proximity theorem is given, which provides a bound that measures how close feasible predictors can come to the optimal predictor. Again, the bound has the asymptotic form (K/2)log n, showing that forecasting trends is fundamentally more difficult than forecasting stationary time series, even when the correct form of the model for the trends is known.

Time Series Simulation with Quasi Monte Carlo Methods

This paper compares quasi Monte Carlo methods, in particular so-called (t, m, s)-nets, with classical Monte Carlo approaches for simulating econometric time-series models. Quasi Monte Carlo methods have found successful application in many fields, such as physics, image processing, and the evaluation of finance derivatives. However, they are rarely used in econometrics. Here, we apply both traditional and quasi Monte Carlo simulation methods to time-series models that typically arise in macroeconometrics. The numerical experiments demonstrate that quasi Monte Carlo methods outperform traditional ones for all models we investigate.

Common large innovations across nonlinear time series

We propose a multivariate nonlinear econometric time series model, which can be used to examine if there is common nonlinearity across economic variables. The model is a multivariate censored latent effects autoregression. The key feature of this model is that nonlinearity appears as separate innovation-like variables. Common nonlinearity can then be easily defined as the presence of common innovations. We discuss representation, inference, estimation and diagnostics. We illustrate the model for US and Canadian unemployment and find that US innovation variables have an effect on Canadian unemployment, and not the other way around, and also that there is no common nonlinearity across the unemployment variables.

Temporal Association, the Dynamics of Crime, and their Economic Determinants: A Time Series Econometric Model of South Africa

The paper examines the association between variouscrime series for South Africa and their economicdeterminants between 1960 and 1993. The study usesJohansen cointegration techniques to establish theimportance of income per capita (economicopportunities), police officers per capita, convictionrates (both of which proxy for the likelihood ofcriminal success whilst the latter also captures aninstitutional efficiency aspect), and politicalinstability (necessary to reflect the unique SouthAfrican environment during this time) on per capitacrime levels. The results caution previous studiesthat find a close relationship between the level ofeconomic activity and recorded crime. Previous workhas relied on the Engle-Granger technique assuming aunique cointegrating vector. Our results for SouthAfrica indicate that whilst total offences arenegatively associated with income per capita,disaggregated crime series do not always yieldcointegrating vectors. Some conclusions on the policyimplications also follow.

Consistent model specification tests for time series econometric models

In this paper we consider general hypothesis testing problems for nonparametric and semiparametric time-series econometric models. We apply the general methodology to construct a consistent test for omitted variables and a consistent test for a partially linear model. The proposed tests are shown to have asymptotic normal distributions under their respective null hypotheses. We also discuss the problems of testing portfolio conditional mean-variance efficiency and testing a semiparametric single index model. Monte Carlo simulations are conducted to examine the finite sample performances of the nonparametric omitted variable test and the test for a partially linear specification.

Static, Dynamic, and Hybrid Neural Networks in Forecasting Inflation

The back-propagation neural network (BPN) model has been the most popular form of artificial neural network model used for forecasting, particularly in economics and finance. It is a static (feed-forward) model which has a learning process in both hidden and output layers. In this paper, we compare the performance of the BPN model with that of two other neural network models, i.e., radial basis function network (RBFN) model and recurrent neural network (RNN) model, in the context of forecasting inflation. The RBFN model is a hybrid model whose learning process is much faster than the BPN model and able to generate almost the same results as the BPN model. The RNN model is a dynamic model which allows feedback from other layers to input layer, enabling it to capture the dynamic behavior of the series. The results of the ANN models are also compared with those of the econometric time series models.

On Seasonal Cycles, Unit Roots, and Mean Shifts

The interpretation of seasonality in terms of economic behavior depends on the form of the econometric time-series model that allows for a description of seasonality. Popular models often assume ei...

Testing the long-run response in time series environmetric and econometric models

Time series models are a common tool in both environmetrics and econometrics. Dynamic effects are accommodated by the inclusion of lagged variables. Tests of null hypotheses involving 'long-run' effects are not invariant and, surprisingly, the most commonly used form involves a non-linear hypothesis when an equivalent linear form is available. As examples we study a simple hydrological model of river flow, a simple electricity demand model and a number of artificially constructed data sets designed to illustrate the issues in a Monte Carlo context. We conclude that the differences due to test form may be large and (a) when it makes a difference, the linear form is to be preferred in terms of accuracy of test size; (b) even with the linear form, with trended data the size distortion may be so large that it may be necessary to simulate critical values for each problem; and (c) if the persistence in the dependent variable is large, test power can be very low.

Testing the long‐run response in time series environmetric and econometric models

Testing the long‐run response in time series environmetric and econometric models

SOME PROPERTIES OF THE MAXIMUM LIKELIHOOD ESTIMATOR IN THE SIMULTANEOUS SWITCHING AUTOREGRESSIVE MODEL

Abstract. The simultaneous switching autoregressive (SSAR) model proposed by Kunitomo and Sato (A non‐linearity in economic time series and disequilibrium econometric models. In Theory and Application of Mathematical Statistics (ed. A. Takemura). Tokyo:University of Tokyo Press (in Japanese), 1994; Asymmetry in economic time series and simultaneous switching autoregressive model. Struct. Change Econ. Dyn., forthcoming (1994).) is a Markovian non‐linear time series model. We investigate the finite sample as well as the asymptotic properties of the least squares estimator and the maximum likelihood (ML) estimator. Due to a specific simultaneity involved in the SSAR model, the least squares estimator is badly biased. However, the ML estimator under the assumption of Gaussian disturbances gives reasonable estimates.

A Bayesian analysis of exogeneity in models pooling time-series and cross-sectional data

The analysis of exogeneity in econometric time-series models as formalized in the seminal paper by Engle et al. [ Econometrica 51 (1983), 277–304] is extended to cover a more general class of models, including error-components models. The Bayesian framework adopted here allows us to take full advantage of a number of statistical tools, related to the reduction of Bayesian experiments, and motivates a careful consideration of prediction issues, leading to a concept of predictive exogeneity. We also adapt the formal definitions of weak and strong exogeneity introduced in Engle et al. (1983), and provide a naturally nested set of definitions for exogeneity. An example highlights the main implications of our analysis for econometric modelling.

The hospital response to Medicare's Prospective Payment System: an econometric model of Blue Cross and Blue Shield plans.

This paper presents an empirical analysis of the impact that resulted from phase-in of Medicare's Prospective Payment System (PPS) on hospital utilization and payments for the Blue Cross and Blue Shield (BCBS) plans. A pooled cross-sectional time series econometric model was specified and estimated using quarterly hospital utilization and payments of the BCBS plans over the period 1980 to 1987. The results indicate that the implementation of PPS was significantly associated with a lower rate of hospital admissions, days and deflated inpatient payments for the BCBS plan members under age 65. A 1% increase in the proportion of hospital days reimbursed under PPS resulted in a .032% decrease in BCBS plan admissions per 1,000 members, a 0.017% decline in days per 1,000 members and a 0.016% decline in deflated inpatient payment per 1,000 members. The reductions in hospital utilization resulted in lower payments by BCBS plans to participating hospitals suggesting a positive spill-over effect of PPS for private insurers. This research underscores the importance of interaction between federal health policy and the private health insurance market.

Macroeconomic forecasting experience with balanced state space models

The paper surveys the forecasting experience in macroeconomics with recently developed balanced state space methods. After relating the state space approach to traditional time series modeling, we give a brief overview of the alternative methods to balanced state space modeling. Results of various recent macroeconomic studies applying these techniques and comparing their forecasting performance to the ones of econometric time series models are discussed.

Testing the recession theory as an explanation for the migration turnaround.

"In this paper the so-called recession theory explanation for the decline of net migration to large metropolitan core areas of industrialized countries is tested with an econometric time-series model. In the explanation it is contended that the migration turnaround represents only a temporary fluctuation in the general trend of urban economic and demographic spatial concentration, caused by the business cycle downturns of the 1970s. Our results show that the migration turnaround cannot be attributed exclusively to these business cycle fluctuations. For many of the countries tested, the business cycle operated simultaneously with other factors suggested as explanations for the turnaround. We conclude that several explanations should be combined to build a theory of the migration turnaround."

Hypothesis Testing in Semiparametric and Nonparametric Models for Econometric Time Series

A restriction on a semiparametric or nonparametric econometric time series model determines the value of a finite-dimensional functional τ of an infinite-dimensional nuisance function. The estimate of τ and its estimated covariance matrix use nonparametric probability and spectral density estimation. A consequent test of the restriction is given approximate large sample justification under absolute regularity on the time series and other conditions. The methodology relates closely to recent proposals of Powell, Stock, Stoker and Robinson in cross-sectional applications, but serial dependence generally affects the test statistic's form, as well as statistical theory.

Why Did the Bank of Canada Emerge in 1935?

Three possible explanations for the emergence of the Canadian central bank in 1935 are examined: that it reflected the need of competitive banking systems for a lender of last resort, that it was necessary to anchor the unregulated Canadian monetary system after abandonment of the gold standard in 1929, and that it was a response to political rather than purely economic pressures. Evidence from a variety of sources (contemporary statements to a Royal Commission, correspondence of chartered bankers, newspaper reports, academic writings, and estimation of time series econometric models) rejects the first two hypotheses and supports the third.

A Standard Aggregative Model for the Jordanian Economy

The simple model presented in this paper is an econometric time-series model which was designed to use the available Jordanian national accounts statistics. It aims at explaining the structural changes in the Jordanian economy in the 1970s as well as projecting values of certain macroeconomic variables for the year 1985, that being the terminal year in the current 5-year plan.

A Statistical Analysis of the Demand for Visits to U.S. National Parks: Travel Costs and Seasonality

The relationship between visits to national parks, seasonality, travel costs, and income is explored through an econometric time series model, covering the years 1965-1981, using monthly data. Two different variables representing travel cost are used: retail gasoline prices and a private transportation index.