Econometric Institute Research Papers
http://repub.eur.nl/col/445/
List of Publicationsenhttp://repub.eur.nl/eur_signature.png
http://repub.eur.nl/
RePub, Erasmus University RepositoryConsensus forecasters: How good are they individually and why?
http://repub.eur.nl/pub/78774/
Mon, 12 Oct 2015 00:00:01 GMT<div>Ph.H.B.F. Franses</div><div>N. Maassen</div>
We analyze the monthly forecasts for annual US GDP growth, CPI inflation rate and the unemployment rate delivered by forty professional forecasters collected in the Consensus database for 2000M01-2014M12. To understand why some forecasters are better than others, we create simple benchmark model-based forecasts. Evaluating the individual forecasts against the model forecasts is informative for how the professional forecasters behave. Next, we link this behavior to forecast performance. We find that forecasters who impose proper judgment to model-based forecasts also have highest forecast accuracy, and hence, they do not perform best just by luck.An Iterative Framework for Real-time Railway Rescheduling
http://repub.eur.nl/pub/78719/
Mon, 05 Oct 2015 00:00:01 GMT<div>T.A.B. Dollevoet</div><div>D. Huisman</div><div>L.G. Kroon</div><div>L.P. Veelenturf</div><div>J.C. Wagenaar</div>
Since disruptions in railway networks are inevitable, railway operators and infrastructure managers need reliable measures and tools for disruption management. Current literature on railway disruption management focuses most of the time on rescheduling one resource (timetable, rolling stock or crew) at the time. In this research, we describe an iterative framework in which all three resources are considered. The framework applies existing models and algorithms for rescheduling the individual resources. We extensively test our framework on instances from Netherlands Railways and show that schedules which are feasible for all three resources can be obtained within short computation times. This shows that the framework and the existing rescheduling approaches can be of great value in practice.Stochastic levels and duration dependence in US unemployment
http://repub.eur.nl/pub/78710/
Wed, 23 Sep 2015 00:00:01 GMT<div>B. de Bruijn</div><div>Ph.H.B.F. Franses</div>
We introduce a new time series model that can capture the properties of data as is typically exemplified by monthly US unemployment data. These data show the familiar nonlinear features, with steeper increases in unem- ployment during economic downswings than the decreases during economic prosperity. At the same time, the levels of unemployment in each of the two states do not seem fixed, nor are the transition periods abrupt. Finally, our model should generate out-of-sample forecasts that mimic the in-sample properties. We demonstrate that our new and flexible model covers all those features, and our illustration to monthly US unemployment data shows its merits, both in and out of sample.Research Ideas for the Journal of Health & Medical Economics: Opinion
http://repub.eur.nl/pub/78715/
Tue, 01 Sep 2015 00:00:01 GMT<div>C-L. Chang</div><div>M.J. McAleer</div>
The purpose of this Opinion article is to discuss some ideas that might lead to papers that are suitable for publication in the Journal of Health and Medical Economics. The suggestions include the affordability and sustainability of universal health care insurance, monitoring and managing costs associated with public and private health and medical care coverage, panel data models based on industrial organization and corporate finance, and health and medical investment finance.Research Ideas for the Journal of Informatics and Data Mining: Opinion
http://repub.eur.nl/pub/78716/
Tue, 01 Sep 2015 00:00:01 GMT<div>M.J. McAleer</div>
The purpose of this Opinion article is to discuss some ideas that might lead to papers that are suitable for publication in the Journal of Informatics and Data Mining. The suggestions include the analysis of citations databases, PI-BETA (Papers Ignored – By Even The Authors), model specification and testing, pre-test bias and data mining, international rankings of academic journals based on citations, international rankings of academic institutions based on citations and other factors, and case studies in numerous disciplines in the sciences and social sciences.Behavioural, Financial, and Health & Medical Economics: A Connection
http://repub.eur.nl/pub/78718/
Tue, 01 Sep 2015 00:00:01 GMT<div>C-L. Chang</div><div>M.J. McAleer</div><div>W-K. Wong</div>
This Opinion article briefly reviews some of the literature in behavioural and financial economics that are related to health & medical economics. We then discuss some of the research on behavioural and financial economics that could be extended to health & medical economics beyond the existing areas in theory, statistics and econometrics.A Column Generation Approach for Locating Roadside Clinics in Africa based upon Effectiveness and Equity
http://repub.eur.nl/pub/78708/
Sat, 15 Aug 2015 00:00:01 GMT<div>J. Núñez Ares</div><div>H. de Vries</div><div>D. Huisman</div>
Long distance truck drivers in Sub-Saharan Africa are extremely vulnerable to HIV and other infectious diseases. The NGO North Star Alliance aims to alleviate this situation by placing so-called Roadside Wellness Centers (RWCs) at busy truck stops along major truck routes. Currently, locations for new RWCs are chosen so as to maximize the expected patient volume and to ensure continuity of access along the routes. As North Star's network grows larger, the objective to provide equal access to healthcare along the different truck routes gains importance. This paper considers the problem to locate a fixed number of RWCs based on these effectiveness and equity objectives. We come up with a novel, set-partitioning type of formulation for the problem and propose a column generation algorithm to solve it. Additionally, we propose and analyze several state-of-the-art acceleration techniques, including dual stabilization, column pool management, and accelerated pricing, which solves the pricing problem as a sequence of shortest path problems. Though the facility location problem is strongly NP-hard, our algorithm yields near-optimal solutions to large randomly generated problem instances within an acceptable amount of time. Our analysis of the trade-off between the equity criterion and North Star's current criteria shows that solutions that are close to optimal with respect to each of the effectiveness and equity objectives are likely to be attainable.Industrial Agglomeration and Use of the Internet
http://repub.eur.nl/pub/78714/
Sat, 01 Aug 2015 00:00:01 GMT<div>C-L. Chang</div><div>M.J. McAleer</div><div>Y-C. Wu</div>
Taiwan has been hailed as a world leader in the development of global innovation and industrial clusters for the past decade. This paper investigates the effects of industrial agglomeration on the use of the internet and internet intensity for Taiwan manufacturing firms, and analyses whether the relationships between industrial agglomeration and total expenditure on internet usage for industries are substitutes or complements. The sample observations are based on 153,081 manufacturing plants, and covers 26 2-digit industry categories and 358 geographical townships in Taiwan. The Heckman selection model is used to adjust for sample selectivity for unobservable data for firms that use the internet. The empirical results from two-stage estimation show that: (1) for the industry overall, a higher degree of industrial agglomeration will not affect the probability that firms will use the internet, but will affect the total expenditure on internet usage; and (2) for 2-digit industries, industrial agglomeration generally decreases the total expenditure on internet usage, which suggests that industrial agglomeration and total expenditure on internet usage are substitutes.Daily Market News Sentiment and Stock Prices
http://repub.eur.nl/pub/78713/
Tue, 28 Jul 2015 00:00:01 GMT<div>D.E. Allen</div><div>M.J. McAleer</div><div>A.K. Singh</div>
In recent years there has been a tremendous growth in the influx of news related to traded assets in international financial markets. This financial news is now available via print media but also through real-time online sources such as internet news and social media sources. The increase in the availability of financial news and investor’s ease of access to it has a potentially significant impact on market price formation as these news items are swiftly transformed into investors sentiment which in turn drives prices. Various commercial agencies have started developing their own financial news data sets which are used by investors and traders to support their algorithmic trading strategies. Thomson Reuters News Analytics (TRNA)1 is one such data set. In this study we use the TRNA data set to construct a series of daily sentiment scores for Dow Jones Industrial Average (DJIA) stock index component companies. We use these daily DJIA market sentiment scores to study the influence of financial news sentiment scores on the stock prices of these companies using a multi-factor model. We use an augmented Fama French Three Factor Model to evaluate the additional effects of financial news sentiment on stock prices in the context of this model. Our results suggest that even when market factors are taken into account, sentiment scores have a significant effect on Dow Jones constituent company returns and that lagged daily sentiment scores are often significant, suggesting that information compounded in these scores is not immediately reflected in security prices and related return series.Multivariate Volatility Impulse Response Analysis of GFC News Events
http://repub.eur.nl/pub/78711/
Wed, 01 Jul 2015 00:00:01 GMT<div>D.E. Allen</div><div>M.J. McAleer</div><div>R.J. Powell</div><div>A.K. Singh</div>
This paper applies the Hafner and Herwartz (2006) (hereafter HH) approach to the analysis of multivariate GARCH models using volatility impulse response analysis. The data set features ten years of daily returns series for the New York Stock Exchange Index and the FTSE 100 index from the London Stock Exchange, from 3 January 2005 to 31 January 2015. This period captures both the Global Financial Crisis (GFC) and the subsequent European Sovereign Debt Crisis (ESDC). The attraction of the HH approach is that it involves a novel application of the concept of impulse response functions, tracing the effects of independent shocks on volatility through time, while avoiding typical orthogonalization and ordering problems. Volatility impulse response functions (VIRF) provide information about the impact of independent shocks on volatility. HH’s VIRF extends a framework provided by Koop et al. (1996) for the analysis of impulse responses. This approach is novel because it explores the effects of shocks to the conditional variance, as opposed to the conditional mean. HH use the fact that GARCH models can be viewed as being linear in the squares, and that multivariate GARCH models are known to have a VARMA representation with non-Gaussian errors. They use this particular structure to calculate conditional expectations of volatility analytically in their VIRF analysis. A Jordan decomposition of Σt is used to obtain independent and identically distributed innovations. A general issue in the approach is the choice of baseline volatilities. VIRF is defined as the expectation of volatility conditional on an initial shock and on history, minus the baseline expectation that conditions on history. This makes the process endogenous, but the choice of the baseline shock within the data set makes a difference. We explore the impact of three different shocks, the first marking the onset of the GFC, which we date as 9 August 2007 (GFC1). This began with the seizure in the banking system precipitated by BNP Paribas announcing that it was ceasing activity in three hedge funds that specialised in US mortgage debt. It took a year for the financial crisis to come to a head, but it did so on 15 September 2008, when the US government allowed the investment bank Lehman Brothers to go bankrupt (GFC2). The third shock is 9 May 2010, which marked the point at which the focus of concern switched from the private sector to the public sector. A further contribution of this paper is the inclusion of leverage, or asymmetric effects. Our modelling is undertaken in the context of a multivariate GARCH model featuring pre-whitened return series, which are then analysed using both BEKK and diagonal BEKK models with the t-distribution. A key result is that the impact of negative shocks is larger, in terms of the effects on variances and covariances, but shorter in duration, in this case a difference between three and six months, in the context of the return series.Axiomatic Characterization of the Median and
Antimedian Function on a Complete Graph minus a
Matching
http://repub.eur.nl/pub/78348/
Thu, 18 Jun 2015 00:00:01 GMT<div>M. Changat</div><div>D.S. Lekha</div><div>S. Mohandas</div><div>H.M. Mulder</div><div>A.R. Subhamathi</div>
__Abstract__
A median (antimedian) of a profile of vertices on a graph G is a vertex that minimizes (maximizes) the sum of the distances to the elements in the profile. The median (antimedian) function has as output the set of medians (antimedians) of a profile. It is one of the basic models for the location of a desirable (obnoxious) facility in a network. The median function is well studied. For instance it has been characterized axiomatically by three simple axioms on median graphs. The median function behaves nicely on many classes of graphs. In contrast the antimedian function does not have a nice behavior on most classes. So a nice axiomatic characterization may not be expected. In this paper an axiomatic characterization is obtained for the median and antimedian function on complete graphs minus a matching.An ABC-Problem for Location and Consensus
Functions on Graphs
http://repub.eur.nl/pub/78320/
Mon, 08 Jun 2015 00:00:01 GMT<div>F.R. McMorris</div><div>B. Novick</div><div>H.M. Mulder</div><div>R.C. Powers</div>
__Abstract__
A location problem can often be phrased as a consensus problem or a voting problem. We use these three perspectives, namely location, consensus and voting to initiate the study of several questions. The median function Med is a location/consensus function on a connected graph G that has the finite sequences of vertices of G as input. For each such sequence, Med returns the set of vertices that minimize the distance sum to the elements of the sequence. The median function satisfies three intuitively clear axioms: (A) Anonymity, (B) Betweenness and (C) Consistency. In [13] it was shown that on median graphs these three axioms actually characterize Med. This result raises a number of questions: (i) On what other classes of graphs is Med characterized by (A), (B) and (C)? (ii) If some class of graphs has other ABC-functions besides Med, then determine additional axioms that are needed to characterize Med. (iii) In the latter case, can we find characterizations of other functions that satisfy (A), (B) and (C)? We call these questions, and related questions, the ABC-Problem for location/consensus functions on graphs. In this paper we present first results. For the first question we use consensus terminology. We construct a non-trivial class different from the median graphs, on which the median function is the unique “ABC function”. For the second and third question voting terminology is most apt for our approach. On K_n with n > 2 we construct various non-trivial ABC-voting procedures. For some nice families, we present a full axiomatic characterization. We also construct an infinite family of ABC-functions on K_3.Volatility Spillovers Between Energy and Agricultural Markets:
A Critical Appraisal of Theory and Practice
http://repub.eur.nl/pub/78349/
Mon, 01 Jun 2015 00:00:01 GMT<div>C-L. Chang</div><div>Y. Li</div><div>M.J. McAleer</div>
__Abstract__
Energy and agricultural commodities and markets have been examined extensively, albeit separately, for a number of years. In the energy literature, the returns, volatility and volatility spillovers (namely, the delayed effect of a returns shock in one asset on the subsequent volatility or covolatility in another asset), among alternative energy commodities, such as oil, gasoline and ethanol across different markets, have been analysed using a variety of univariate and multivariate models, estimation techniques, data sets, and time frequencies. A similar comment applies to the separate theoretical and empirical analysis of a wide range of agricultural commodities and markets. Given the recent interest and emphasis in bio-fuels and green energy, especially bio-ethanol, which is derived from a range of agricultural products, it is not surprising that there is a topical and developing literature on the spillovers between energy and agricultural markets. Modelling and testing spillovers between the energy and agricultural markets has typically been based on estimating multivariate conditional volatility models, specifically the BEKK and DCC models. A serious technical deficiency is that the Quasi-Maximum Likelihood Estimates (QMLE) of a full BEKK matrix, which is typically estimated in examining volatility spillover effects, has no asymptotic properties, except by assumption, so that no statistical test of volatility spillovers is possible. Some papers in the literature have used the DCC model to test for volatility spillovers. However, it is well known in the financial econometrics literature that the DCC model has no regularity conditions, and that the QMLE of the parameters of DCC has no asymptotic properties, so that there is no valid statistical testing of volatility spillovers. The purpose of the paper is to evaluate the theory and practice in testing for volatility spillovers between energy and agricultural markets using the multivariate BEKK and DCC models, and to make recommendations as to how such spillovers might be tested using valid statistical techniques. Three new definitions of volatility and covolatility spillovers are given, and the different models used in empirical applications are evaluated in terms of the new definitions and statistical criteria.Symbolic Multidimensional Scaling
http://repub.eur.nl/pub/78189/
Wed, 27 May 2015 00:00:01 GMT<div>P.J.F. Groenen</div><div>Y. Terada</div>
__Abstract__
Multidimensional scaling (MDS) is a technique that visualizes dissimilarities between pairs of objects as distances between points in a low dimensional space. In symbolic MDS, a dissimilarity is not just a value but can represent an interval or even a histogram. Here, we present an overview of developments for symbolic MDS. We discuss how interval dissimilarities they can be represented by (concentric) circles or rectangles, how replications can be represented by a three-way MDS version, and show how nested intervals of distances can be obtained for representing histogram dissimilarities. The various models are illustrated by empirical examples.How to gain brain for Suriname
http://repub.eur.nl/pub/78151/
Thu, 21 May 2015 00:00:01 GMT<div>T.W. Dulam</div><div>Ph.H.B.F. Franses</div>
__Abstract__
This paper investigates whether high skilled migrants of Surinamese origin would be willing to return to the home country if they were offered a remigration benefits package. We surveyed 209 highly educated individuals of Surinamese origin who live in the Netherlands. A quarter of them is willing to return to Suriname if they were offered a house, land property, and easy access to credit. Eliminating political interference in profession would even attract the majority. The willingness to accept the offer diminishes over time. The offer mostly attracts engineers to return to Suriname. Offering funds for research and innovation attracts health professionals as well. We also explore some other proposals and discuss the policy implications.International Technology Diffusion of Joint
and Cross-border Patents
http://repub.eur.nl/pub/78143/
Fri, 15 May 2015 00:00:01 GMT<div>C-L. Chang</div><div>M.J. McAleer</div><div>J-T. Tang</div>
__Abstract__
With the advent of globalization, economic and financial interactions among countries
have become widespread. Given technological advancements, the factors of production
can no longer be considered to be just labor and capital. In the pursuit of economic
growth, every country has sensibly invested in international cooperation, learning,
innovation, technology diffusion and knowledge. In this paper, we use a panel data set
of 40 countries from 1981 to 2008 and a negative binomial model, using a novel set of
cross-border patents and joint patents as proxy variables for technology diffusion, in
order to investigate such diffusion. The empirical results suggest that, if it is desired to
shift from foreign to domestic technology, it is necessary to increase expenditure on
R&D for business enterprises and higher education, exports and technology. If the
focus is on increasing bilateral technology diffusion, it is necessary to increase
expenditure on R&D for higher education and technology.Risk attitudes in the board room and company performance:
Evidence for an emerging economy
http://repub.eur.nl/pub/78127/
Fri, 01 May 2015 00:00:01 GMT<div>D.S. Bodeutsch</div><div>Ph.H.B.F. Franses</div>
__Abstract__
We personally interview thirteen board members of seven (out of the ten) companies listed at the
Suriname Stock Exchange and ask questions about their past and current decisions and on their
risk attitudes. Next, we correlate the answers to company performance in between 2003-2011, like
earnings per share, stock returns, book value and market value. Recent literature on risk attitudes
in the board, which usually draws on western economies, guides our formulation of hypotheses.
At the same time we also perform some exploratory analyses. Our main result is that, for this
emerging economy, more risk adversity leads to better firm performance.A Stochastic Dominance Approach to the Basel III Dilemma: Expected Shortfall or VaR?
http://repub.eur.nl/pub/78155/
Fri, 01 May 2015 00:00:01 GMT<div>C-L. Chang</div><div>J.A. Jiménez-Martín</div><div>M.J. McAleer</div><div>T. Pérez-Amaral</div>
__Abstract__
The Basel Committee on Banking Supervision (BCBS) (2013) recently proposed shifting the quantitative risk metrics system from Value-at-Risk (VaR) to Expected Shortfall (ES). The BCBS (2013) noted that “a number of weaknesses have been identified with using VaR for determining regulatory capital requirements, including its inability to capture tail risk” (p. 3). For this reason, the Basel Committee is considering the use of ES, which is a coherent risk measure and has already become common in the insurance industry, though not yet in the banking industry. While ES is mathematically superior to VaR in that it does not show “tail risk” and is a coherent risk measure in being subadditive, its practical implementation and large calculation requirements may pose operational challenges to financial firms. Moreover, previous empirical findings based only on means and standard deviations suggested that VaR and ES were very similar in most practical cases, while ES could be less precise because of its larger variance. In this paper we find that ES is computationally feasible using personal computers and, contrary to previous research, it is shown that there is a stochastic difference between the 97.5% ES and 99% VaR. In the Gaussian case, they are similar but not equal, while in other cases they can differ substantially: in fat-tailed conditional distributions, on the one hand, 97.5%-ES would imply higher risk forecasts, while on the other, it provides a smaller down-side risk than using the 99%-VaR. It is found that the empirical results in the paper generally support the proposals of the Basel Committee.Least-squares Bilinear Clustering of Three-way Data
http://repub.eur.nl/pub/77757/
Mon, 16 Feb 2015 00:00:01 GMT<div>P.C. Schoonees</div><div>P.J.F. Groenen</div><div>M. van de Velden</div>
__Abstract__
A least-squares bilinear clustering framework for modelling three-way data, where each observation consists of an ordinary two-way matrix, is introduced. The method combines bilinear decompositions of the two-way matrices into overall means, row margins, column margins and row-column interactions with clustering along the third way. Different clusterings are defined for each part of the decomposition, so that up to four different classifications are defined jointly. The computational burden is greatly reduced by the orthogonality of the bilinear model, such that the joint clustering problem reduces to separate ones which can be handled independently. Three of these sub-problems are specific cases of $k$-means clustering; a special algorithm is formulated for the row-column interactions, which are displayed in clusterwise biplots. The method is illustrated via two empirical examples and interpreting the interaction biplots are discussed.On the Invertibility of EGARCH(p,q)
http://repub.eur.nl/pub/78126/
Sun, 01 Feb 2015 00:00:01 GMT<div>G.G. Martinet</div><div>M.J. McAleer</div>
__Abstract__
Of the two most widely estimated univariate asymmetric conditional volatility models,
the exponential GARCH (or EGARCH) specification can capture asymmetry, which
refers to the different effects on conditional volatility of positive and negative effects of
equal magnitude, and leverage, which refers to the negative correlation between the
returns shocks and subsequent shocks to volatility. However, the statistical properties of
the (quasi-) maximum likelihood estimator (QMLE) of the EGARCH parameters are not
available under general conditions, but only for special cases under highly restrictive and
unverifiable conditions, such as EGARCH(1,0) or EGARCH(1,1), and possibly only
under simulation. A limitation in the development of asymptotic properties of the QMLE
for the EGARCH(p,q) model is the lack of an invertibility condition for the returns
shocks underlying the model. It is shown in this paper that the EGARCH(p,q) model can
be derived from a stochastic process, for which the invertibility conditions can be stated
simply and explicitly. This will be useful in re-interpreting the existing properties of the
QMLE of the EGARCH(p,q) parameters.