Econometrics and Beyond Celebrating 60 Years of the Econometric Institute and 50 Years of FAECTOR May 27, 2016
Econometrics and Beyond
1
Ladies and gentlemen,
Great to see you all! Welcome to the Erasmus School of Economics, and in particular, welcome to our Econometric Institute. This year is a remarkable year, as our Institute celebrates its 12th lustrum, but even better, our vibrant student association FAECTOR celebrates its 50 years of existence. Worldwide, FAECTOR is the oldest and largest econometrics students association, and we are very proud of what our students have achieved (and still do today).
A
nd, what is the best way to celebrate a birthday? Yes, that is to organise a conference and invite alumni, former and current colleagues and friends. That we are here today shows a key feature of Rotterdam-grown econometricians, and that is that we have a large network of befriended fellow former class mates, who work and live everywhere around the world. We all share a common background, and when we meet, a few words are enough. This state of mind, when added to the quality of the Rotterdam students, makes working and teaching here at the Econometric Institute a great pleasure.
2
Econometrics and Beyond
Personally I will not make it to the first centennial meeting of FAECTOR in 2066, but I reckon that all recent graduates and current students will. That will be very spectacular! I wish you all a very pleasant day at our campus, with an inspiring conference, and a lively meeting with fellow econometricians, who all look back at a few very important years of their lives. Enjoy this day! Prof. dr. Philip Hans Franses Dean of the Erasmus School of Economics Econometrics and Beyond
3
Programme 13:00 – 13:30
Registration (C-hall, Theil Building)
13:30 – 14:15
oom CB-1: Opening lecture by Prof. dr. Rinnooy Kan R Chaired by Prof. dr. Patrick Groenen (Director Econometric Institute)
14:15 – 15:15
Parallel sessions: oom CB-2: Instrumental Variables R Chaired by Prof. dr. Dennis Fok oom CB-5: Operations Research, Informatics and Beyond I R Chaired by Prof. dr. Dennis Huisman oom CB-1: Entrepreneurship in Econometrics I R Chaired by Luc Oudenes (President FAECTOR)
15:15 – 15:45
Coffee break
15:45 – 17:15
Parallel sessions:
Session
Instrumental Variables and Score Driven Models
oom CB-2: Score Driven Models R Chaired by Prof. dr. Dick van Dijk oom CB-5: Operations Research, Informatics and Beyond II R Chaired by Prof. dr. Albert Wagelmans oom CB-1: Entrepreneurship in Econometrics II R Chaired by Luc Oudenes (President FAECTOR) Session:
Instrumental Variables
Room:
CB-2
oom Theatre at Pavilion: Closing lecture by Prof. dr. Philip Hans Franses R Chaired by Luc Oudenes (President FAECTOR)
14:15 – 14:45
Frank Kleibergen
14:45 – 15:15
Jan Groen
18:00 – 19:00
Reception at Erasmus Pavilion
Session:
Score Driven Models
Room:
CB-2
19:00
Alumni & EI Dinner at the Erasmus Pavilion
15:45 – 16:15
André Lucas
16:15 – 16:45
Peter Schotman
16:45 – 17:15
Anne Opschoor
17:15 – 17:30 17:30 – 18:00
4
Break and move to Pavilion
Econometrics and Beyond
Econometrics and Beyond
5
Session: Instrumental Variables Time: 14:15 – 14:45 Room: CB-2 Speaker: Prof. dr. Frank Kleibergen (University of Amsterdam) Title: Weak Identification
T
raditional two step estimation methods like two stage least squares to estimate linear instrumental variable regression models and Fama-MacBeth two pass regressions to estimate linear factor models suffer from weak identification issues. Weak identification occurs when the explanatory power in the first step regression is low which commonly occurs. Weak identification robust testing procedures have been developed which render size correct tests in all instances. These tests have similar power as conventional tests but remain trustworthy when identification is weak. I will provide an overview of currently existing procedures and discuss remaining open issues.
About Frank Kleibergen Frank Kleibergen is currently professor of econometrics at the University of Amsterdam. He was previously professor of economics at Brown University. Frank Kleibergen is renowned for his work on weak identification but also worked extensively in other fields like Bayesian econometrics and time series analysis. He has supervised a considerable number of graduate students some of which are well known in the field.
6
Econometrics and Beyond
Session: Instrumental Variables Time: 14:45 – 15:15 Room: CB-2 Speaker: Dr. Jan Groen (Federal Reserve Bank of New York) Title: Estimating the Impact of Oil Shocks on the U.S. Economy
I
f oil prices reflect demand shocks, then an oil price decline, say, would suggest a slowdown in global economic activity. Alternatively, if the oil prices are driven by supply shocks, then the drop in prices might indicate a forthcoming boost in spending as firms and households benefit from lower energy costs. In this paper, I use correlations of oil price changes with a broad array of financial variables to decompose oil price changes in components due to global oil demand and those due to global oil supply. As the latter is a true oil shock and not so much an endogenous response as is the case for demand drive oil price changes, I use the identified supply price component as an instrumental variable in a structural VAR model to assess how oil supply shocks affect U.S. economic conditions.
About Jan Groen Jan J. J. Groen is an Officer in the Research & Statistics Group at the Federal Reserve Bank of New York, specializing in international finance and econometric modeling. Prior to joining the New York Fed he operated in a number of senior economist roles at the Bank of England, was a Research Fellow at the Dutch Central Bank, and served as a member of the Econometric Institute at the Erasmus University Rotterdam. Dr. Groen’s research has been published in high ranking academic journals and covers topics such as empirical exchange rate modeling, international finance and macroeconomics, and forecasting and econometric modeling. He has been visiting a scholar and consultant at several foreign research and policy institutions, and he was a member of the editorial board of the Journal of Money, Credit, and Banking. Dr. Groen graduated with a PhD in Economics from the Tinbergen Institute at the Erasmus University Rotterdam.
Econometrics and Beyond
7
Session: Score Driven Models Time: 15:45 – 16:15 Room: CB-2 Speaker: Prof. dr. André Lucas (VU Amsterdam) Title: Scoring in Econometrics with Score Models
O
ver the last 8 years, we have developed a new flexible class of time series models. These bring together all the goodies we have been trained for and that we naturally like as econometricians: thinking about new models and building an intuition for them, crossfertilization between optimization (OR) and time series (Econometrics), general umbrellas for many well-known models as well as unexpected discoveries of new models, hardcore asymptotics and tangible simulation results. This talk provides a brief tour on this new class of models for nonexperts and meanwhile touches on some anecdotes about the process of scientific discovery and dissemination.
About André Lucas André Lucas is a professor of finance at VU Amsterdam. He received his M.Sc. and his PhD in Econometrics from Erasmus University Rotterdam and has published on financial econometrics and time series econometrics as well as risk management in journals such as the Journal of Econometrics, Journal of Applied Econometrics, Journal of Business and Economic Statistics, Review of Economics and Statistics, Journal of Financial and Quantitative Analysis, and more. Together with a number of co-authors he propagates the use of generalized autoregressive score dynamics for time varying parameter models (see www.gasmodel.com). Over 2010-2015 he received a five year prestigious VICI research grant for this project from the Dutch national research council (NWO).
8
Econometrics and Beyond
Session: Score Driven Models Time: 16:15 – 16:45 Room: CB-2 Speaker: Prof. dr. Peter Schotman (Maastricht University) Title: Score-Driven Nelson Siegel: Hedging Long-Term Liabilities
D
ue to its affine structure the Nelson-Siegel model for yield curves can be transformed to a factor model for excess bond returns. Hedging interest rate risk in this framework amounts to eliminating the factor exposure and minimizing the residual risk. Fitting the model directly on excess returns with constant factor loadings leads to large hedging errors caused by substantial and persistent time-variation in the shape parameter of the Nelson-Siegel factor loadings. To capture this variation we develop a Dynamic Conditional Score (DCS) model for the shape parameter. This dynamic model offers superior hedging performance and reduces the hedging error standard deviation by almost 50% during the financial crisis. Much of the improvement is due to the model for the shape parameter with some further reduction achieved by a GARCH model for the residual risk.
About Peter Schotman Peter Schotman is Professor of Empirical Finance at Maastricht University School of Business and Economics. He studied econometrics at Erasmus University Rotterdam, where he also obtained his PhD at the Econometric Institute. His current research focuses on long term investments in relation to retirement provision, both from the perspective of pension funds as well as households. Most of his work uses methods from financial econometrics. He is a senior research fellow of Netspar (Network for Studies on Pensions, Aging and Retirement) and often works together with practitioners in the pension industry.
Econometrics and Beyond
9
Session: Score Driven Models Time: 16:45 – 17:15 Room: CB-2 Speaker: Dr. Anne Opschoor (VU Amsterdam) Title: Fractionally Integrated Multivariate Models for Fat-Tailed Realized Covariance Kernels and Returns
W
e introduce a new fractionally integrated model for multivariate covariance matrix dynamics based on the long-memory behaviour of daily realized covariance matrix kernels and daily return observations. We account for fat-tailedness in both types of data by assuming a matrix-F distribution for the realized kernels and a multivariate Student’s t distribution for the returns. In addition, the score-driven propagation mechanism adopted for the covariance matrix dynamics endows our model with further robustness properties. Using intraday stock data over the period 2001-2012, we construct realized covariance kernels and show that the new fractionally integrated model outperforms recent alternatives such as the Multivariate HEAVY model and the Riskmetrics 2006 (long-memory) model both statistically and economically.
About Anne Opschoor Anne Opschoor (1987) obtained his master’s degree in econometrics with honors from Erasmus University Rotterdam in November 2009. In that same month he started as a PhD student at the Tinbergen Institute and Econometric Institute at the Erasmus University, supervised by TI fellows professor Dick van Dijk and Michel van der Wel. He defended his thesis ‘Understanding Financial Market Volatility’ in February 2014 and won the 2014 Journal of Applied Econometrics Dissertation Prize for his paper “Improving Density Forecasts and Value-at-Risk Estimates by Combining Densities”. Anne published in the Journal of Econometrics, Journal of Statistical Software and Journal of Empirical Finance and is co-author of the book “Time Series Models for Business and Economic Forecasting’’ (2nd Edition), written with professor Philip Hans Franses and professor Dick van Dijk. At the moment, he is an assistant professor of finance at Vrije Universiteit Amsterdam. His primary fields of interest are modeling financial time series, with a particular focus on volatility modeling and forecasting.
10
Econometrics and Beyond
Session
Operations Research, Informatics and Beyond
Session:
Operations Research, Informatics and Beyond I
Room:
CB-5
14:15 – 14:45
Ruud Teunter
14:45 – 15:15
Robin Nicolai
Session:
Operations Research, Informatics and Beyond II
Room:
CB-5
15:45 – 16:15
Edwin Romeijn
16:15 – 16:45
Ludo Waltman
16:45 – 17:15
Hans Frenk
Econometrics and Beyond
11
Session: Operations Research, Informatics and Beyond I Time: 14:15 – 14:45 Room: CB-5 Speaker: Prof. dr. Ruud Teunter (University of Groningen) Title: Inventory Control: Selecting the Right Service Level for Each Stock Keeping Unit
A
ll most all firms stock thousands of different Stock Keeping Units (SKUs), and is seems quite obvious that these should not all be treated in the same way. In particular, the customer service level should vary across the SKUs. However, the large number of SKUs makes it difficult to deal with this manually and, surprisingly, existing inventory software offers little assistance. “System based” approached have been suggested in the literature, but their complexity may explain their limited success in being adopted. In my presentation, I will present simpler approaches that I have developed. First, I will explain why existing ABC classification approaches should not be used for setting service levels at the SKU level, and propose an alternative but equally simple classification method. Next, I will let go of classification altogether and present an even simpler and intuitive way of varying service levels. Results on real life datasets show a savings potential of up to 30% on inventory investment. The latest method has also been adopted in a major inventory software package and has been implemented at two companies.
About Ruud Teunter Ruud Teunter obtained his PhD in Groningen and then worked for about 7 years as a Post-Doc researcher together with Rommert Dekker, first in several countries within a European project on reverse logistics led by Rommert and later at the EI in Rotterdam. He then spent five years at Lancaster University in the UK, after which he closed the loop and returned to Groningen where he now heads the Operations Department. He has published over 60 papers on operations management and research in journals such as Production and Operations Management, European Journal of Operational Research, Manufacturing and Service Operations Management, and the Journal of Operations Management. He also publishes in professional outlets such as Logistiek.nl and Maintenance Magazine. His research interests cover maintenance and service logistics, reverse logistics, forecasting and inventory control. He is currently the leading partner for the NWO/ProRail funded research project PYRAMIDS on maintenance optimization for ProRail, and the project leader for two projects with academic and business partners on service logistics for the process industries (€ 1.6M) and on service logistics for offshore windmills (€ 1.3M).
12
Econometrics and Beyond
Session: Operations Research, Informatics and Beyond I Time: 14:45 – 15:15 Room: CB-5 Speaker: Dr. Robin Nicolai (HKV Consultants) Title: A Necessary Condition for Flood Safety in the Netherlands
I
t is often claimed that The Netherlands is the safest Delta in the world. Protection against flooding is a vital issue since at least 55% of the country is vulnerable to flooding. Each year the Dutch government spends roughly 1 billion Euros on protection by dikes and dunes. In total there are 3,500 kilometers of primary flood defences in the Netherlands. The statutory flood protection standards for these primary flood defences date back to the early 50’s. After the 1953 flood (in Dutch: Watersnoodramp) the first Delta committee asked mathematician David van Dantzig to determine economically optimal dike heights. His research was published in Econometrica in 1960 and his recommendations formed the basis for the current safety standards. Almost 60 years later the population in The Netherlands has grown to 17 million people and the economic value of the low-lying areas behind dikes (Dutch: polders) has increased substantially. Moreover, sea-level rise and sinking land adversely affect the country’s flood risk. A national flood risk assessment showed that it was time to reconsider the protection levels. Recently, the Delta program introduced new safety standards for the primary flood defence system using a risk-based approach. The approach takes into account both the probability of flooding and the consequences related to flooding. The consequences are expressed in terms of casualties and economic losses. With respect to economic losses a large-scale cost-benefit analysis and optimization has been performed. In 2013 the ‘Dutch Delta team’ that solved the problem “Given cost constraints, how can the Netherlands best improve flood protection while minimizing the total long-term expected costs?” earned the Franz Edelman Award in operations research and management science. It is clear that Operations Research techniques play a prominent role in optimizing flood protection in The Netherlands. Robin shall give a short overview of studies and their results on this topic in recent years, including the most challenging OR-issues. Robin shall also demonstrate the use of OR in other flood risk management problems. For example, flood risk assessments require the tail distribution of sea-water levels, storm surges, discharges and precipitation levels, which can be found by applying extreme value theory. Historical records on these natural phenomena are mostly limited to 100-150 years, yielding large statistical uncertainties in water levels with return periods higher than a thousand years. Since we have not measured such low-frequency water levels in The Netherlands, the physical modelling of water levels inhibits large uncertainties as well. The question is how to avoid the accumulation of such uncertainties in e.g. estimating flood damage and longterm decision-making on flood protection measures.
Econometrics and Beyond
13
Another research topic Robin shall address is the cost-optimal maintenance and replacement of hydraulic structures in the waterways and bodies of water in The Netherlands. The last decades the use of the comprehensive and aging infrastructure has intensified. Decision-making on maintenance requires lifetime estimates of structures and information on maintenance/ replacement costs in course of time. Robin has developed and applied Bayesian methods to lifetime data of large hydraulic structures and concrete bridges. The resulting lifetime distributions were used to calculate long-term budget requirements for the replacement of these structures. Many (future) OR challenges remain in the field of flood risk management. One can think of cost-effective strategies for replacing storm surge barriers given sea-level rise and economic developments, the role of uncertainties in (operational) flood risk management, data management, drought and water distribution issues, and dredging optimization.
About Robin Nicolai Robin Nicolai (1979) studied Econometrics at Erasmus University Rotterdam (EUR). In 2002 he obtained his MSc degree with a thesis on the optimization of a simulation model for breastcancer screening. In 2008 he obtained his PhD degree from EUR with a thesis on maintenance optimization. Since 2008 he works at HKV Consultants (Lelystad), a company that provides advisory services, high quality research and software products in the field of water and safety. In his role as advisor, researcher and project leader he applies his knowledge of mathematical modelling and statistics in flood risk assessments, uncertainty analyses of hydraulic loads on flood defences and maintenance management of hydraulic structures in The Netherlands.
14
Econometrics and Beyond
Session: Operations Research, Informatics and Beyond II Time: 15:45 – 16:15 Room: CB-5 Speaker: Prof. dr. Edwin Romeijn (Georgia Tech) Title: Quantifying the Trade-Off Between Radiation Therapy Treatment Plan Quality and Delivery Efficiency
B
eam-on-time is an important measure of the delivery efficiency in Intensity-Modulated Radiation Therapy (IMRT). Traditionally, minimizing beam-on-time has been postponed until the Leaf Sequencing stage where the treatment plan quality is already determined and fixed. However, there is a trade-off between the beam-on-time and the treatment plan quality. The aim of this study is to incorporate the beam-on-time into the treatment-plan optimization stage using a Direct Aperture Optimization approach. This will allow for explicitly quantifying that trade-off. In particular, using the special structure of the problem, an exact solution approach is proposed which sequentially characterizes segments of the Pareto-efficient frontier. Furthermore, an approximate solution method, applicable to more classes of treatment plan evaluation criteria, is developed. This solution method is tested on clinical cancer cases and its performance is compared to general approximation techniques available for convex bi-criteria optimization problems.
About Edwin Romeijn Edwin Romeijn received his M.S. (1988) and PhD (1992) degrees from Erasmus University Rotterdam in The Netherlands. He has been a faculty member at the Rotterdam School of Management, the University of Florida, and the Department of Industrial and Operations Engineering at the University of Michigan, as well as a Program Director at the National Science Foundation for the Operations Research, Manufacturing Enterprise Systems, and Service Enterprise Systems programs. Since January 2015 he serves as the H. Milton and Carolyn J. Stewart School Chair of the H. Milton Stewart School of Industrial and Systems Engineering at Georgia Tech. He has taught courses in operations research, stochastic processes, applied probability and statistics, supply chain management, and decision support systems. His research focuses on optimization theory and applications, in particular in the areas of supply chain optimization and optimization in health care.
Econometrics and Beyond
15
Session: Operations Research, Informatics and Beyond II Time: 16:15 – 16:45 Room: CB-5 Speaker: Dr. Ludo Waltman (Centre for Science and Technology Studies) Title: From Econometrics to Bibliometrics
I
will give an overview of my research program on advanced bibliometric methods at the Centre for Science and Technology Studies (CWTS) of Leiden University. This research program is aimed at developing quantitative methods for extracting information from the meta-data of scientific publications. The research program is strongly focused on network analysis, big data analysis, and visualization. I will discuss the techniques and tools that have been developed, including the popular VOSviewer software tool. The development of this tool started originally at the Econometric Institute. I will also discuss how bibliometric research benefits from ideas originating from the fields of economics and econometrics.
About Ludo Waltman Ludo Waltman is a researcher at the Centre for Science and Technology Studies (CWTS) of Leiden University. He is responsible for the research program on advanced bibliometric methods, and he is co-developer of two widely used bibliometric software tools: VOSviewer (www.vosviewer.com) and CitNetExplorer (www.citnetexplorer.nl). Ludo is Editor-in-Chief of Journal of Informetrics, one of the core journals in the field of bibliometrics and scientometrics. Ludo obtained his PhD at the Econometric Institute in 2011.
16
Econometrics and Beyond
Session: Operations Research, Informatics and Beyond II Time: 16:45 – 17:15 Room: CB-5 Speaker: Prof. dr. Hans Frenk (Sabanci University) Title: On Marked Poisson Point Processes and Martingales and their Application to Airline Revenue Management and Inventory Control
I
n this short talk we will discuss the application of the theory of marked Poisson point processes and martingales to the overbooking problem in airline revenue management and two problems in inventory control. The purpose of this talk is to give an example of the application of some theory of stochastic processes normally not taught in an industrial engineering graduate program to derive in an efficient way the objective function for these type of finite horizon problems. At the same time it is also possible to incorporate into the modeling some important features related to the arrival process of customers. These features cannot be analyzed by the classical techniques from time homogeneous Markov processes and Markov chains since these techniques mainly deal with deriving average case results.
About Hans Frenk Hans Frenk joined the Faculty of Engineering and Natural Sciences at Sabanci University, Istanbul, 7 years ago. He graduated from Utrecht University (master degree in mathematics) and did his PhD thesis (1983) within the field of renewal theory and regenerative processes at the Econometric Institute, Erasmus University, Rotterdam, The Netherlands under the supervision of Prof. dr. L.F.M de Haan. Through the years his main research interest is in the theory of stochastic processes and optimization. At the moment his research is in applications of those techniques to problems in Management Science and Engineering. (maintenance, inventory control and revenue management).
Econometrics and Beyond
17
Session
Entrepreneurship in Econometrics
Session: Entrepeneurship in Econometrics I Time: 14:15 – 14:45 Room: CB-1 Speaker: Karsten Uil Start-up: Charta Software
Summary arsten is cofounder and managing director of Charta Software. He studied Econometrics between 1996 and 2001 at the Erasmus University and graduated in the direction of logistics under supervision of Rommert Dekker. With this background he gained working experience at ORTEC and TNO before he started Charta Software with his partner Peter-Jan Roes (Econometrics / Besliskunde) in 2006.
K
Start-up “Since 2000, Peter-Jan and I discussed the flaws in software development, we talked and coded until late at night and philosophized about how software should be developed. With my consultancy and commercial edge and Peter-Jan’s ability to build anything efficiently, one day we decided: Let’s just do it. We quit our jobs and the company Charta Software was a fact.” Charta Software Charta Software has grown to a specialized company of eight mathematicians and software developers. The software provides solutions for all kinds of complex processes in (mainly, but not restricted to) the logistic and financial domain. It is used by dozens of companies and hundreds of users every day.
Session:
Entrepreneurship in Econometrics I
Room:
CB-1
14:15 – 14:45
Karsten Uil
14:45 – 15:15
Sem Kippers
Session:
Entrepreneurship in Econometrics II
Room:
CB-1
15:45 – 16:15
Edgar Mul
16:15 – 16:45
Robbert Bos
16:45 – 17:15
Gerrit Timmer
18
Econometrics and Beyond
Econometrics and Beyond
19
Session: Entrepreneurship in Econometrics I Time: 14:45 – 15:15 Room: CB-1 Speaker: Sem Kippers Start-up: Jedlix
I
n 2015 together with another colleague Mr. Kippers founded Jedlix, the Smart charging solution of Eneco to charge Electric vehicles smartly regarding sustainability and price. In a short period, Jedlix achieved becoming the smart charging partner for Renault and other car manufactures will follow shortly. Mr. Kippers is currently responsible for designing and implementing the new infrastructure to facilitate Eneco’s ambition to become an important party as an electricity flexibility provider to balance the electricity grid’s in Europe. In this talk, Mr. Kippers shares his insights in the advantages of being an econometrician in embracing disruptive thinking and acting. About Sem Kippers Sem Kippers studied Econometrics at the Erasmus University between 2008 and 2012. After that he started at the trading department at Eneco as a forecast analyst.
Session: Entrepreneurship in Econometrics II Time: 15:45 – 16:15 Room: CB-1 Speaker: Edgar Mul Start-up: NEWCRAFT
About Edgar Mul Edgar Mul studied Econometrics at Erasmus University and graduated with an MSc degree in 1999, after which he started working at ECT as a marketeer. He then moved on to work as a programme director and management consultant for some of the Netherlands largest companies (Shell, Philips, KPN, to name a few). In 2006, he joined the Management team of Ordina, and later became the e-commerce manager for Thomas Cook Netherlands. He left this post to found Conversion Company, an e-commerce accelerator with the objective of helping businesses maximize their online results. In 2014, Mul started a new company, NEWCRAFT, aimed at aiding companies understand their customer’s digital attitude.
20
Econometrics and Beyond
Session: Entrepreneurship in Econometrics II Time: 16:15 – 16:45 Room: CB-1 Speaker: Robbert Bos Start-up: Veneficus
I
n his talk, Robbert will focus on how he is building a company around factual decision making. He will speak about his vision on development within data science. Will it be possible to automatize our work?
About Robbert Bos As one of the founders of Veneficus Robbert has built a data analytics firm. He started his study Econometrics in 2003. During his final Bachelor year he founded Veneficus along with three of his classmates. After his bachelor he completed a masters in Quantitative Marketing to broaden his horizons within the profession while working at Veneficus. His company Veneficus is a research and consulting company, which is committed to support businesses in different markets with (big) data, analytics and visualization. They are telling the story of data. As a partner of Veneficus he is still involved with many of the company’s projects and is focused on the mission of building bridges between different disciplines such as marketing, finance and IT.
Session: Entrepreneurship in Econometrics II Time: 16:45 – 17:15 Room: CB-1 Speaker: Prof. dr. Gerrit Timmer (VU Amsterdam) Start-up: ORTEC
About Gerrit Timmer Gerrit Timmer studied econometrics at the Erasmus University from 1974 to 1979. In 1984, he received his PhD for his research on stochastic methods for global optimization under the inspirational supervision of Alexander Rinnooy Kan. Mr. Timmer never completely left the academic world. Currently, he is full professor in business econometrics at the Free University in Amsterdam (part time). Together with some fellow student from the Erasmus University, he founded ORTEC in 1981. Based on modelling, simulation, and optimization techniques, ORTEC develops and implements products that increases performance and sustainable growth for its customers. He is active as a Board member of ORTEC, which has a 1000 employees worldwide including many econometricians. In his presentation, Gerrit Timmer will focus on various aspects of “how to create a business based on OR”. He will summarize his experience in a number of lessons learned when applying OR in practice.
Econometrics and Beyond
21
Acknowledgement We would like express our thanks to the sponsors of this event. • The Econometric Institute • Erasmus Research Institute of Management • Erasmus School of Economics • Tinbergen Institute
22
Econometrics and Beyond
Econometrics and Beyond
23
ERASMUS UNIVERSITY ROTTERDAM CAMPUS WOUDESTEIN
ROUTE Public Transportation From Rotterdam Central Station • Take trams 21 and 24 heading to De Esch, get off at stop Woudestein. • With the metro heading to De Akkers or Slinge, change at Beurs on to the metro towards De Terp, Nesselande or Binnenhof, get off at stop Kralingse Zoom. From Rotterdam Alexander • With the metro heading to Schiedam Centrum or De Akkers, get off at stop Kralingse Zoom.
24
Econometrics and Beyond
By car From highway A16, take exit 25 to Centrum (sign). From there at the roundabout, follow Centrum (Abram van Rijckevorselweg). Then, at the first traffic lights, turn right. Turn right again to enter the Erasmus University Parking lot. Address Erasmus University Rotterdam Campus Woudestein Burgemeester Oudlaan 50 | 3062 PA Rotterdam E: conference@faector.nl