Cointegrated Time Series Research Articles

Do non-intermediation services tell us more in the finance–growth nexus?: causality evidence from eight OECD countries

ABSTRACTThis paper provides new evidence on a traditional finance–growth nexus through dividing financial services into financial intermediation and non-intermediation services and examining their relationships with economic growth. Applying time-series cointegration techniques and Granger causality tests for eight Organization for Economic Cooperation and Development (OECD) countries, reveals several results. First, there is a long-run equilibrium relationship among economic growth, intermediation activities, and non-intermediation activities in Austria, France, and Korea. Second, non-intermediation services impede long-run economic growth in Austria and France, whereas non-intermediation business and financial intermediation services accelerate Korea’s long-term growth. Third, weak exogeneity tests support long-run bi-directional causality and the supply-leading hypothesis in terms of the relationship between financial services and economic growth. Finally, the influences of intermediation and non-intermediation activities on economic growth vary across countries, financial services, and time periods, indicating that countries should adopt different financial services to enhance long- or short-term economic growth. This paper emphasizes the importance of non-intermediation activities in the growth process and in the development of intermediation services.

Do Non-Intermediation Services Tell Us More in the Finance–Growth Nexus?: Causality Evidence from Eight OECD Countries

This paper provides new evidence on a traditional finance–growth nexus through dividing financial services into financial intermediation and non-intermediation services and examining their relationships with economic growth. Applying time-series cointegration techniques and Granger causality tests for eight Organization for Economic Cooperation and Development (OECD) countries, reveals several results. First, there is a long-run equilibrium relationship among economic growth, intermediation activities, and non-intermediation activities in Austria, France, and Korea. Second, non-intermediation services impede long-run economic growth in Austria and France, whereas non-intermediation business and financial intermediation services accelerate Korea’s long-term growth. Third, weak exogeneity tests support long-run bi-directional causality and the supply-leading hypothesis in terms of the relationship between financial services and economic growth. Finally, the influences of intermediation and non-intermediation activities on economic growth vary across countries, financial services, and time periods, indicating that countries should adopt different financial services to enhance long- or short-term economic growth. This paper emphasizes the importance of non-intermediation activities in the growth process and in the development of intermediation services.

Is commercial real estate a good hedge against inflation? Evidence from South Africa

This study investigates the inflation hedging ability of Commercial Real Estate (CRE) in South Africa. The Vector Error Correction (VEC) model for cointegrated time series was used to investigate the long run relationship between property returns and inflation and to determine whether inflation drives property returns, and if so, to what degree it drives returns. This study finds that in the short run, CRE investments are generally a pervasive hedge against inflation. In the long run, retail and industrial property hedge against inflation, with retail property being the better inflation hedge of the two property types.

Does unemployment aggravate suicide rates in South Africa? Some empirical evidence

The purpose of our study is to investigate the cointegration relationship between suicides and unemployment in South Africa using annual data collected between 1996 and 2015. The study relies on the bounds approach to autoregressive distributive lag (ARDL) cointegration model to establish long-run and short-run cointegration between unemployment and suicide rates. Furthermore, suicide data is further disintegrated into ‘sex’ and ‘age’ demographics to provide a more disaggregated analysis. Our empirical results indicate that unemployment is insignificantly related with suicide rates with the exception for citizens above 75 years. On the other hand, other control variables such as per capita GDP, inflation and divorce appear to be more significantly related with suicides. To the best of our knowledge, this study becomes the first to investigate the time series cointegration relationship between suicides and unemployment for South Africa data.

Error-Correction Factor Models for High-dimensional Cointegrated Time Series

Cointegration inferences often rely on a correct specification for the short-run dynamic vector autoregression. However, this specification is unknown, a priori. A lag length that is too small leads to an erroneous inference as a result of the misspecification. In contrast, using too many lags leads to a dramatic increase in the number of parameters, especially when the dimension of the time series is high. In this paper, we develop a new methodology which adds an error-correction term for the long-run equilibrium to a latent factor model in order to model the short-run dynamic relationship. The inferences use the eigenanalysis-based methods to estimate the cointegration and latent factor process. The proposed error-correction factor model does not require an explicit specification of the short-run dynamics, and is particularly effective for high-dimensional cases, in which the standard error-correction suffers from overparametrization. In addition, the model improves the predictive performance of the pure factor model. The asymptotic properties of the proposed methods are established when the dimension of the time series is either fixed or diverging slowly as the length of the time series goes to infinity. Lastly, the performance of the model is evaluated using both simulated and real data sets.

Improving control efficiency in buffering systems using anticipatory indicators for demand forecasting

Optimization of the stock management process is associated with the search for a forecasting model, a method for generation of a forecasting time series, a model of logistic operation, determining a reasonable level of reserve stocks and establishing the optimization criterion.Successful solution to the optimization problem in general can be achieved only if the whole complex of local management problems is successfully solved. In this case, the method of generation of a cointegrated time series of demand forecasting is the central element of the technology of optimal stocks management. This relates to the fact that probabilistic nature of demand is the main factor reducing efficiency of management in systems of this class.It was shown that the proposed method for improving management efficiency can be used in any economic system due to the possibility of construction of a single logistic operation model.The proposed approach is based on formation of a time series specifically designed to solve the problem of forecasting the demand in stocks buffering systems. Such a series contains both information on sales volumes and data related to consumer demand.Since consumer activity is ahead of the process of physical consumption of products, it becomes possible to use anticipatory markers in forecasting problems.The study of operational processes using a verified indicator of efficiency has confirmed the hypothesis of presence of anticipatory markers within the framework of the formed forecast time series.It has been established that the maximum management efficiency can be achieved in the case of a lower construction accuracy of the forecast model. This is due to the fact that the logistic operation model takes into account the costs of movement of products and their valuation at the operation input and output.

Natural disasters and Malaysian economic growth: policy reforms for disasters management.

The natural catastrophic events largely damage the country's sustainability agenda through massive human fatalities and infrastructure destruction. Although it is partially supported the economic growth through the channel of "Schumpeter creative destruction" hypothesis, however, it may not be sustained in the long-run. This study examined the long-run and causal relationships between natural disasters (i.e., floods, storm, and epidemic) and per capita income by controlling FDI inflows and foreign aid in the context of Malaysia, during the period of 1965-2016. The study employed time series cointegration technique, i.e., autoregressive distributed lag (ARDL)-bounds testing approach for robust inferences. The results show that flood, storm, and epidemic disasters substantially decrease the country's per capita income, while FDI inflows and foreign aid largely supported the country's economic growth in the short-run. These results are disappeared in the long-run, where flood and storm disasters exhibit the positive association with the economic growth to support the Schumpeter creative destruction hypothesis. The foreign aid decreases the per capita income and does not maintain the "aid-effectiveness" hypotheses in a given country. The causality estimates confirmed the disaster-led growth hypothesis, as the causality estimates running from (i) storm to per capita income, (ii) epidemic to per capita income, and (iii) storm to foreign aid. The results emphasized for making disaster action plans to reduce human fatalities and infrastructure for sustainable development.

Modelos alternativos para explicar la evolución del gasto público español

El artículo contrasta empíricamente diferentes modelos alternativos, con variables tanto desde el lado de la oferta como de la demanda, para estudiar la evolución del gasto público en España en el período 1964-2016. Se utiliza la metodología habitual de raíces unitarias y cointegración en series temporales para contrastar los modelos, utilizando como base modelos de retardo distribuido autorregresivo para el largo plazo y modelos de corrección de errores para el corto. De los tres modelos considerados, el que mejor explica el comportamiento del sector público español es el modelo de hacienda pública al que se le ha añadido la variable explicativa de la burocracia.

Granger Causality Testing in Mixed‐Frequency VARs with Possibly (Co)Integrated Processes

We analyze Granger causality (GC) testing in mixed‐frequency vector autoregressions (MF‐VARs) with possibly integrated or cointegrated time series. It is well known that conducting inference on a set of parameters is dependent on knowing the correct (co)integration order of the processes involved. Corresponding tests are, however, known to often suffer from size distortions and/or a loss of power. Our approach works for MF variables that are stationary, integrated of an arbitrary order, or cointegrated. As it only requires the estimation of a MF‐VAR in levels with appropriately adjusted lag length, after which GC tests can be conducted using simple standard Wald tests, it is of great practical appeal. In addition, we show that the presence of non‐stationary and trivially cointegrated high‐frequency regressors leads to standard distributions when testing for causality on a subset of parameters, sometimes even without any need to augment the VAR order. Monte Carlo simulations and two applications involving the oil price and consumer prices as well as GDP and industrial production in Germany illustrate our approach.

Is Commercial Real Estate a Good Hedge Against Inflation? Evidence From South Africa

This study investigate the inflation hedging ability of Commercial Real Estate (CRE) in South Africa. The Vector Error Correction (VEC) model for cointegrated time series was used to investigate the long run relationship between property returns and inflation and to determine whether inflation drives property returns, and if so, to what degree it drives returns. This study finds that in the short run, CRE investments are generally a pervasive hedge against inflation. In the long run, retail and industrial property hedge against inflation, with retail property being the better inflation hedge of the two property types.

NonSTOP: A NonSTationary Online Prediction Method for Time Series

We present online prediction methods for time series that let us explicitly handle nonstationary artifacts (e.g., trend and seasonality) present in most real time series. Specifically, we show that applying appropriate transformations to such time series before prediction can lead to improved theoretical and empirical prediction performance. Moreover, since these transformations are usually unknown, we employ the learning with experts setting to develop a fully online method (NonSTOP-NonSTationary Online Prediction) for predicting nonstationary time series. This framework allows for seasonality and/or other trends in univariate time series and cointegration in multivariate time series. Our algorithms and regret analysis subsume recent related work while significantly expanding the applicability of such methods. For all the methods, we provide sublinear regret bounds. We support all of our results with experiments on simulated and real data.

Determinants of international tourism demand: Evidence from Australian states and territories

This article explores whether the determinants of international tourism demand differ by states and territories in Australia. This is the first attempt at econometric modelling of international tourism demand in the states and territories of Australia. A demand model is specified where international visits to states and territories is a function of world income, state-level transportation costs, stock of foreign-born residents, the Australian real exchange rate and the price levels of international and domestic substitutes. Panel and time series econometric techniques are employed to test the model variables for stationarity, cointegration and direction of causality. Panel and time series cointegration tests show that the model is cointegrated. The causality analysis indicates that all explanatory variables Granger cause international visits to the Australian states and territories. Further, we show that the impacts of the determinants of international tourism vary by states and territories. The results underscore the importance of targeted policymaking that takes into account the economic and social structure of each state and territory instead of designing tourism policies on the basis of one-size-fits-all approach.

Cointegration Analysis among the Variables of the Ohlson Model for Brazilian Companies

We examine whether there is a long-term equilibrium relation between the companies market value and the variables accounting book value and abnormal earnings based on the Ohlson model (1995) using a cointegration approach. Our panel cointegration analysis indicates that the variables cointegrate when using the whole sample, the most liquid companies group and for all sectors in at least one of the tests performed with exception of the Telecommunications sector, which presented no cointegration in both tests. The time series cointegration results have shown that, except for one company, for all the remaining the variables cointegrated. Therefore, the Ohlson Model (1995) is relevant for the evaluation of Brazilian listed companies in a long-term equilibrium. In addition, we provide evidence that abnormal earnings have limited explanatory power compared to book value.

Are the Economies of Canada and the United States Integrated? Evidence from Cointegration Analysis

ABSTRACTIn recognition of his contributions to the development of the method of cointegration analysis for analyzing nonstationary time series, late Sir Clive William John Granger (September 4, 1934–May 27, 2009) was awarded the Nobel Memorial Prize in Economic Sciences in 2003. Since then, this method has become a dominant paradigm in empirical economic research. However, this method is not without critics. This article is one in a series to point out some inconsistent arguments used in the development of the method of cointegration analysis. To illustrate by example, we apply the method of time series cointegration analysis and present statistical evidence that supports the proposition that the economies of Canada and the United States are integrated. We conclude this article by laying out a foundation to formally criticize the method of cointegration analysis in subsequent research.

Assessing interdependence among countries' fundamentals and its implications for exchange rate misalignment estimates: An empirical exercise based on GVAR

Abstract Exchange rates are important macroeconomic prices and changes in these rates affect economic activity, prices, interest rates, and trade flows. Methodologies have been developed in empirical exchange rate misalignment studies to evaluate whether a real effective exchange is overvalued or undervalued. There is a vast body of literature on the determinants of long-term real exchange rates and on empirical strategies to implement the equilibrium norms obtained from theoretical models. This study seeks to contribute to this literature by showing that the global vector autoregressions model (GVAR) proposed by Pesaran and co-authors can add relevant information to the literature on measuring exchange rate misalignment. Our empirical exercise suggests that the estimative exchange rate misalignment obtained from GVAR can be quite different to that using the traditional cointegrated time series techniques, which treat countries as detached entities. The differences between the two approaches are more pronounced for small and developing countries. Our results also suggest a strong interdependence among eurozone countries, as expected.

Residential Electricity Use Effects of Population in Kazakhstan

We studied impacts of population groups of 15-64 and 65-above on residential electricity use in Kazakhstan in the STIRPAT framework. Unlike earlier studies for Kazakhstan in the STIRPAT framework, we applied time series cointegration and error correction methods. Results from the autoregressive distributed lags bounds testing approach indicate a significant impact of the age group of 15-64 on the residential electricity use in long-run, however, the age group of 65-above has only short-run effects and affluence has no effect. Another finding is that, 21% of short-run disequilibrium can be corrected towards long-run equilibrium during a year. Policymakers should consider the trend of the population group of 15-64 in their decision about the long-run stance of the residential electricity consumption. The trend suggests an implementation of energy conservative policy and increasing efficiency of its usage. Another policy implication is that household's electricity consumption is not income dependent maybe due to cheap electricity prices subsidised by the government. In the short-run, policy makers should consider the age group of 65-above among other factors in their implementations. Moreover, they should be careful in making any policy shock to the residential electricity consumption system, because convergence towards long-run equilibrium path takes about six years.

Time Series Analysis for Spatial Node Selection in Environment Monitoring Sensor Networks.

Wireless sensor networks are widely used in environmental monitoring. The number of sensor nodes to be deployed will vary depending on the desired spatio-temporal resolution. Selecting an optimal number, position and sampling rate for an array of sensor nodes in environmental monitoring is a challenging question. Most of the current solutions are either theoretical or simulation-based where the problems are tackled using random field theory, computational geometry or computer simulations, limiting their specificity to a given sensor deployment. Using an empirical dataset from a mine rehabilitation monitoring sensor network, this work proposes a data-driven approach where co-integrated time series analysis is used to select the number of sensors from a short-term deployment of a larger set of potential node positions. Analyses conducted on temperature time series show 75% of sensors are co-integrated. Using only 25% of the original nodes can generate a complete dataset within a 0.5 °C average error bound. Our data-driven approach to sensor position selection is applicable for spatiotemporal monitoring of spatially correlated environmental parameters to minimize deployment cost without compromising data resolution.

Long-term regional house prices cycles. A city-based index for Italy

PurposeThis paper aims to present the dynamics of housing prices in Italian cities based on unpublished data with regional details from the late 1960s, half-yearly base, for all main Italian cities measuring the average prices for three city dimensions: city centre, sub-centres and outskirts or suburbs. It estimates the Italian long-term house price index, city based in real terms, and shows a combination of methods to deal with large time-series data.Design/methodology/approachThis paper builds long-term cycles based on the city (real) data by estimating the common components of cointegrated time series and extracting the unobservable signals to build real house price index for sub-regions in Italy. Three different econometric methodologies are used: Johansen cointegration test and VAR models to identify the long-term pattern of prices at the estimated aggregate level; principal components to obtain the common (permanent and transitory) components; and signal extraction in ARIMA time series–model-based approach method to extract the unobserved time signals.FindingsResults show three long-term cycle-trends during the period and identify several one-direction causal non-permanent relationships among house prices from different Italian areas. There is no evidence of convergence among regional’s house prices suggesting that the Italian housing prices converge inside the local market with only short diffusion effects at larger regional level.Research limitations/implicationsData are measured as the average price in squared meters, and the resulting index is not quality controlled.Practical implicationsThe long-term trends on housing prices serve to implement further research and know deeply the evolution of Italian housing prices.Originality/valueThis paper contains new and unknown information about the evolution of housing prices in Italian regions and cities.

Understanding price discovery in interconnected markets: Generalized Langevin process approach and simulation

While the literature on price discovery process and information flow between dominant and satellite market is exhaustive, most studies have applied an approach that can be traced back to Hasbrouck (1995) or Gonzalo and Granger (1995). In this paper, however, we propose a Generalized Langevin process with asymmetric double-well potential function, with co-integrated time series and interconnected diffusion processes to model the information flow and price discovery process in two, a dominant and a satellite, interconnected markets. A simulated illustration of the model is also provided.

SPECIFICATION TESTING DRIVEN BY ORTHOGONAL SERIES FOR NONLINEAR COINTEGRATION WITH ENDOGENEITY

This paper proposes two simple and new specification tests based on the use of an orthogonal series for a considerable class of bivariate nonlinearly cointegrated time series models with endogeneity and nonstationarity. The first test is proposed for the case where the regression function is integrable, which fills a gap in the literature, and the second test, which nests the first one, deals with regression functions in a quite large function space that is sufficient for both theoretical and practical use. As a starting point of our asymptotic theory, the first test is studied initially and then the theory is extended to the second test. Endogeneity in two general forms is allowed in the models to be tested. The finite sample performance of the tests is examined through several simulated examples. Our experience generally shows that the proposed tests are easily implementable and also have stable sizes and good power properties even when the ‘distance’ between the null hypothesis and a sequence of local alternatives is asymptotically negligible.