Showing papers on "Cointegration published in 1995"

Likelihood-Based Inference in Cointegrated Vector Autoregressive Models

Abstract: This book gives a detailed mathematical and statistical analysis of the cointegrated vector autoregresive model This model had gained popularity because it can at the same time capture the short-run dynamic properties as well as the long-run equilibrium behaviour of many non-stationary time series It also allows relevant economic questions to be formulated in a consistent statistical framework Part I of the book is planned so that it can be used by those who want to apply the methods without going into too much detail about the probability theory The main emphasis is on the derivation of estimators and test statistics through a consistent use of the Guassian likelihood function It is shown that many different models can be formulated within the framework of the autoregressive model and the interpretation of these models is discussed in detail In particular, models involving restrictions on the cointegration vectors and the adjustment coefficients are discussed, as well as the role of the constant and linear drift In Part II, the asymptotic theory is given the slightly more general framework of stationary linear processes with iid innovations Some useful mathematical tools are collected in Appendix A, and a brief summary of weak convergence in given in Appendix B The book is intended to give a relatively self-contained presentation for graduate students and researchers with a good knowledge of multivariate regression analysis and likelihood methods The asymptotic theory requires some familiarity with the theory of weak convergence of stochastic processes The theory is treated in detail with the purpose of giving the reader a working knowledge of the techniques involved Many exercises are provided The theoretical analysis is illustrated with the empirical analysis of two sets of economic data The theory has been developed in close contract with the application and the methods have been implemented in the computer package CATS in RATS as a result of a rcollaboation with Katarina Juselius and Henrik Hansen

Estimation of Common Long-Memory Components in Cointegrated Systems

Abstract: The study of cointegration in large systems requires a reduction of their dimensionality. To achieve this, we propose to obtain the I(1) common factors in every subsystem and then analyze cointegration among them. In this article, a new way of estimating common long-memory components of a cointegrated system is proposed. The identification of these I(1) common factors is achieved by imposing that they be linear combinations of the original variables Xt , and that the error-correction terms do not cause the common factors at low frequencies. Estimation is done from a fully specified error-correction model, which makes it possible to test hypotheses on the common factors using standard chi-squared tests. Several empirical examples illustrate the procedure.

Convergence in international output

Abstract: SUMMARY This paper proposes and tests new definitions of convergence and common trends for per capita output. We define convergence for a group of countries to mean that each country has identical long-run trends, either stochastic or deterministic, while common trends allow for proportionality of the stochastic elements. These definitions lead naturally to the use of cointegration techniques in testing. Using centurylong time series for 15 OECD economies, we reject convergence but find substantial evidence for common trends. Smaller samples of European countries also reject convergence but are driven by a lower number of common stochastic trends.

Using Cointegration Analysis in Econometric Modelling

Abstract: 1. Introduction and Overview 2. Short - and Long-Run Models 3. Testing for Unit Roots 4. Cointegration in Single Equations 5. Cointegration in Multivariate Systems 6. Modelling the Short-Run and Other Extensions Appendix References Index

Fully modified least squares and vector autoregression

Abstract: Fully modified least squares (FM-OLS) regression was originally designed in work by Phillips and Hansen (1990) to provide optimal estimates of cointegrating regressions. The method modifies least squares to account for serial correlation effects and for the endogeneity in the regressors that results from the existence of a cointegrating relationship. This paper provides a general framework which makes it possible to study the asymptotic behavior of FM-OLS in models with full rank I(1) regressors, models with I(1) and I(0) regressors, models with unit roots, and models with only stationary regressors. This framework enables us to consider the use of FM regression in the context of vector autoregressions (VAR's) with some unit roots and some cointegrating relations. The resulting FM-VAR regressions are shown to have some interesting properties. For example, when there is some cointegration in the system, FM-VAR estimation has a limit theory that is normal for all of the stationary coefficients and mixed normal for all of the nonstationary coefficients. Thus, there are no unit root limit distributions even in the case of the unit root coefficient submatrix (i.e., I n-r , for an n-dimensional VAR with r cointegrating vectors). Moreover, optimal estimation of the cointegration space is attained in FM-VAR regression without prior knowledge of the number of unit roots in the system, without pretesting to determine the dimension of the cointegration space and without the use of restricted regression techniques like reduced rank regression. The paper also develops an asymptotic theory for inference based on FM-OLS and FM-VAR regression. The limit theory for Wald tests that rely on the FM estimator is shown to involve a linear combination of independent chi-squared variates. This limit distribution is bounded above by the conventional chi-squared distribution with degrees of freedom equal to the number of restrictions. Thus, conventional critical values can be used to construct valid (but conservative) asymptotic tests in quite general FM time series regressions. This theory applies to causality testing in VAR's and is therefore potentially useful in empirical applications.

Sustainability of the Deficit Process With Structural Shifts

Abstract: This article discusses the condition for deficit sustainability and searches for shifts in the structure of U.S. deficit policy. I show that, contrary to the current literature, cointegration between revenues and expenditures inclusive of debt payment is not a necessary but a sufficient condition for a strict interpretation of deficit sustainability. The necessary condition requires that debt grow slower than the borrowing rate. For tests on structural shifts, I use a test for structural change that searches for shifts in the rank of the cointegrating matrix. I find that there is a shift in deficit policy in the early 80s so that cointegration between revenue and expenditure inclusive of interest payments holds only up to 1980. I show, however, that the deficit process is still sustainable despite the failure of cointegration in the 80s. The finding of shifts in the deficit process is more uncertain when I normalize by gross national product or population, but the deficit is shown to be sustainable regard...

Cointegration, Error Correction, and Price Discovery on Informationally Linked Security Markets

Abstract: Using synchronous transactions data for IBM from the New York, Pacific, and Midwest Stock Exchanges, we estimate an error correction model to investigate whether each of the exchanges is contributing to price discovery. Johansen's test yields two cointegrating vectors, which together verify the expected long-run equilibrium of equal prices across the three exchanges. Two error correction terms specified as the differences from IBM prices on the NYSE indicate that adjustments maintaining the long-run cointegration equilibrium take place on all three exchanges. That is, IBM prices on the NYSE adjust toward IBM prices on the Midwest and Pacific Exchanges, just as Midwest and Pacific prices adjust to the NYSE.

Arbitrage, cointegration, and testing the unbiasedness hypothesis in financial markets

Abstract: We use a no-arbitrage, cost-of-carry asset pricing model to show that the existence of coin? tegration between spot and forward (futures) prices depends on the time-series properties of the cost-of-carry. We argue that the conditions for cointegration are more likely to hold in currency markets than in commodity markets, explaining many of the empirical results in the literature. We also use this model to demonstrate why the forward rate forecast error, the basis, and the forward premium are serially correlated, and to develop econometric tests ofthe "unbiasedness hypothesis" (sometimes called the "simple efficiency hypothesis") in various financial markets. The unbiasedness hypothesis is so prevalent in the finance liter? ature that many tests for it have been developed. We examine four of the common tests and and use our cointegration results to demonstrate why each of these tests should reject the null hypothesis of unbiasedness. We find strong support for our hypothesis in the existing empirical literature.

A Stastistical Analysis of Cointegration for I(2) Variables

Abstract: This paper discusses inference for I(2) variables in a VAR model. The estimation procedure suggested consists of two reduced rank regressions. The asymptotic distribution of the proposed estimators of the cointegrating coefficients is mixed Gaussian, which implies that asymptotic inference can be conducted using the χ2 distribution. It is shown to what extent inference on the cointegration ranks can be conducted using the tables already prepared for the analysis of cointegration of I(1) variables. New tables are needed for the test statistics to control the size of the tests. This paper contains a multivariate test for the existence of I(2) variables. This test is illustrated using a data set consisting of U.K. and foreign prices and interest rates as well as the exchange rate.

An investigation of cointegration and causality between energy consumption and economic growth

Abstract: This paper reexamines the causality between energy consumption and economic growth with both bivariate and multivariate models by applying the recently developed methods of cointegration and Hsiao`s version of the Granger causality to transformed U.S. data for the period 1947-1990. The Phillips-Perron (PP) tests reveal that the original series are not stationary and, therefore, a first differencing is performed to secure stationarity. The study finds no causal linkages between energy consumption and economic growth. Energy and gross national product (GNP) each live a life of its own. The results of this article are consistent with some of the past studies that find no relationship between energy and GNP but are contrary to some other studies that find GNP unidirectionally causes energy consumption. Both the bivariate and trivariate models produce the similar results. We also find that there is no causal relationship between energy consumption and industrial production. The United States is basically a service-oriented economy and changes in energy consumption can cause little or no changes in GNP. In other words, an implementation of energy conservation policy may not impair economic growth. 27 refs., 5 tabs.

Chapter 32 Perspectives on PPP and long-run real exchange rates

Abstract: Publisher Summary This chapter presents an overview of the long-run determinants of purchasing power parity (PPP). It reviews the huge time series literature testing simple PPP. This area has proven fruitful ground for applying modern methods for dealing with nonstationary and near-nonstationary time series. The chapter traces out the evolution of the literature from naive static tests of PPP to modern unit-root approaches for testing whether real exchange rates are stationary and to cointegration techniques—the most recent phase of PPP testing. The research on more disaggregated price data is discussed in the chapter, including a nearly two-hundred year data set on commodity prices in England and France during the seventeenth and eighteenth centuries. Aside from providing an extremely long data set, this historical data offers some perspective on the behavior of cross-country relative prices in more modern times. The chapter looks at some possible medium- and long-run determinants of the real exchange rate, particularly the supply-side determinants emphasized in the popular Balassa–Samuelson model. It also considers some evidence that positive demand shocks, such as unexpected increases in government spending, lead to medium-run appreciations of the real exchange rate.

Long-Run Determinants of the Real Exchange Rate: A Stock-Flow Perspective

Abstract: This paper examines the long-run determinants of the real exchange rate from a stock-flow perspective. The empirical analysis estimates a long-run relationship between the real exchange rate, net foreign assets, and other factors affecting trade flows. Using postwar data for the United States and Japan, cointegration analysis supports the finding that the structural factors underlying each country's net trade and net foreign asset positions determine the long-run path for the real value of the dollar and the yen. The empirical analysis also provides estimates for the underlying stochastic trend in each real exchange rate series.

The causal relationship between taxes and expenditures in the G7 countries: cointegration and error-correction models

Abstract: The paper examines the causal relationship between tax revenues and expenditures in the G7 countries using cointegration and error-correction methodology. This statistical technique provides additional channels through which causality could emerge. The empirical results show that bidirectional causality exists between government taxes and expenditures in all countries except in Japan and Italy. For Japan and Italy, causality runs from government taxes to expenditures.

Modeling Volatility Dynamics

Abstract: Good macroeconomic and financial theorists, like all good theorists, want to get the facts straight before theorizing; hence, the explosive growth in the methodology and application of time-series econometrics in the last twenty-five years. Many factors fueled that growth, ranging from important developments in related fields (see Box and Jenkins, 1970) to dissatisfaction with the “incredible identifying restrictions” associated with traditional macroeconometric models (Sims, 1980) and the associated recognition that many tasks of interest, such as forecasting, simply do not require a structural model (see Granger and Newbold, 1979). A short list of active subfields includes vector autoregressions, index and dynamic factor models, causality, integration and persistence, cointegration, seasonality, unobserved-components models, state-space representations and the Kalman filter, regime-switching models, nonlinear dynamics, and optimal nonlinear filtering. Any such list must also include models of volatility dynamics. Models of autoregressive conditional heteroskedasticity (ARCH), in particular, provide parsimonious approximations to volatility dynamics and have found wide use in macroeconomics and finance1. The family of ARCH models is the subject of this chapter.

The Development of Children Adopted from Romanian Orphanages.

Abstract: A cointegration framework is used to examine the short-run and long-run characteristics of energy demand in the Australian road transport sector. A lagged endogenous equation based on a partial adjustment process is proposed and estimated. Results indicate that energy demand, output and real energy prices are integrated of order 1 and cointegrated. The long-run output and price elasticities of energy demand are estimated to be 0.52 and -0.12 respectively. Causality tests reveal a bidirectional causality path between output and energy demand and a unidirectional path from energy consumption to prices. No other causality paths between output, prices and energy demand are detected. The short-run output elasticity of energy demand is estimated to be 0.25 based on an error-correction model. The short-run price elasticity is found to be insignificant. The inertia parameter is 0.48 corresponding to 95% of the demand adjustment occurring after five periods. The results are compared with previous findings and the variations are partially attributed to the structural changes in the road transport sector in the 1980s, some of which are discussed.

A causality analysis of the link between higher education and economic development

Abstract: The aim of this article is to shed light on the relationship between higher education and economic development by means of econometric tools designed to evaluate the existence and direction of causality: cointegration and Granger-causality tests. The results show a significant causality from national higher educational effort (proxied by the number of students per capita, i.e. not engaged in productive activities) to economic development for four countries: Sweden (1910–1986), United Kingdom (1919–1987), Japan (1885–1975) and France (1899–1986). However, such a causality link has not been found for Italy (1885–1986) or Australia (1906–1986). This suggests that this relationship is indeed not mechanistic as already pointed out by some social scientists.

Estimating common sectoral cycles

Abstract: We investigate in this paper the degree of short-run and long-run comovement in U.S. sectoral output data by estimating sectoral trends and cycles. A theoretical model based on Long and Plosser (1983) is used to derive a reduced form for sectoral outputs from first principles. Cointegration and common-cycle tests are performed; sectoral output data seem to share a relatively high number of common trends and a relatively low number of common cycles. A special trend-cycle decomposition of the data set is performed, and the results indicate a very similar cyclical behavior across sectors and very different behavior for trends. In a variance decomposition analysis, prominent sectors such as Manufacturing and Wholesale/Retail Trade exhibit relatively important transitory shocks.

The causal relationships between equity indices on world exchanges

Abstract: The Engle and Granger cointegration analysis and Granger causality tests are applied to monthly time series of nine major stock market indices over the period January 1982 to February 1991 to examine for causal linkages. The empirical results indicate that there is adequate evidence to refute the notion of informationally efficient stock markets.

Time-varying distributions and the optimal hedge ratios for stock index futures

Abstract: This paper estimates the risk-minimizing futures hedge ratios for three types of stock index futures: S&P 500 index futures, major market index (MMI) futures and Toronto 35 index futures. Spot and futures prices are first analysed to adjust for non-stationarity and cointegration. Using a bivariate cointegration model with a generalized ARCH error structure, we estimate the optimal hedge ratio as a ratio of the conditional covariance between spot and futures to the conditional variance of futures. Both within-sample comparisons and out-of-sample comparisons reveal that the dynamic hedging strategy based on the bivariate GARCH estimation improves the hedging performance over the conventional constant hedging strategy

Modelling Inflation in Australia

Abstract: This paper develops an empirically constant, data-coherent, error correction model for inflation in Australia. The level of consumer prices is a mark-up over domestic and import costs, with adjustments for dynamics and relative aggregate demand. We address issues of cointegration, general to specific modelling, dynamic specification, model evaluation and testing, parameter constancy, and exogeneity. We also test this model against existing models of Australian prices: this model encompasses (but is not encompassed by) the existing models.

The long-run gains from international equity diversification: Australian evidence from cointegration tests

Abstract: The benefits available from international equity diversification to Australian investors for the period 1970–92 are analysed using monthly index data for 16 countries supplied by Morgan Stanley Capital International. The cointegration framework is utilized and results from the standard Engle—Granger two-step ordinary least squares procedure are compared with those from the Johansen (1988) maximum likelihood procedure. It is found that, as in other recent work (Taylor and Tonks, 1989; Andrade, Clare and Thomas, 1991), the two techniques lead to different conclusions in certain cases. It is also found that, as in Kasa (1992), there is evidence of cointegration among a subset of the indices considered. A further finding of interest to the applied worker is that the results in the Johansen procedure are sensitive to the VAR specification and we believe that Hall's (1991) warning regarding the reporting of tests from this procedure is valid.

An integrated analysis of turkish inflation

Abstract: The inflation process in Turkey is investigated utilizing the multivariate cointegration model that is based on the joint analysis of long-run and short-run behavior. Excess demand in each sector--monetary, government, international and labor--as measured by the deviation from the long-run equilibrium is allowed to potentially effect the inflation rate. Excess demand in the government sector is found to be the main determinant of the Turkish inflation rate. Thus, a rate policy implication is that inflation could be reduced rapidly by eliminating the fiscal deficit. Copyright 1995 by Blackwell Publishing Ltd

Cointegration tests of present value models with a time‐varying discount factor

Abstract: The paper analyses the impact of persistence and volatility in the discount rate in present-value models on cointegration tests in levels and in logarithms. In simulations we find that the probability of not rejecting the null of no cointegration depends on the persistence of the discount rate process and can be very high when the expected returns process is highly persistent. In contrast, the cointegration tests are very robust with respect to the level of volatility in the discount rate. We discuss the relevance of our findings for the US stock market where standard ADF tests do not reject the null of no cointegration between stock prices and dividends. Based on estimates of persistence in four asset pricing models, we find that a model which links expected returns to the dividend yield is sufficiently persistent to explain the failure of rejecting the null that stock prices and dividends are not cointegrated.