Sanja Dudukovic
Professor Emerita, International Management
Department Co-Chair, International Management
Ph.D. University of Belgrade, Yugoslavia
M.S. University of Belgrade, Yugoslavia
B.S. University of Belgrade, Yugoslavia
Office: Lowerre Academic Center, North Campus 4
Phone: +41 91 986 36 34
This email address is being protected from spambots. You need JavaScript enabled to view it.
Profile:
Since 1991, Dr. Dudukovic has taught a number of courses in Business Analytics, including Data Mining (Business Intelligence), Management Science, Management Information Systems, Quantitative Methods & Dynamic Forecasting, Statistics and Computing. Her degrees include a B.S. in Technology, an M.S. in Economics and a Ph.D. in Statistics. Her long-term theoretical research interests include Non-Gaussian Time Series Modeling, Entropy Maximization and Information Theory. Since 1997, her empirical research interests cover the fields of Financial Modeling, Credit Spread Modeling, Stock Market GARCH and RV Volatility Forecasting , Machine Learning and HOC ARMA modeling. She has published numerous publications on Non-Gaussian Time Series Analysis and Volatility Forecasting and has considerable private-sector experience in Management Information System Development. She has a particular interest in Educational Data Mining and undergraduate research. She is engaged in discovering the methods needed to trigger student research interest by including notions of creativity in philosophy, psychology and logic. She established and directed Franklin's Center for Quantitative Research (CQR) with the aim of achieving results worth presenting at professional international conferences. She is a member of the Bernoulli Society for Mathematical Statistics, the American Statistical Association and the IEEC Computer Society.
2020-2021 Courses:
| BUS 306 | Quantitative Methods and Dynamic Forecasting | FALL 2020 |
| In the first part of this course students learn concepts in inferential statistics, its main principles and algorithms. They learn how to apply sampling distributions in the case of business random variables, how to state and test business hypotheses about population mean or proportion differences, how to calculate ANOVA table components, and how to deploy estimation methods to provide information needed to solve real business problems. In the second part of the course, students learn advanced model building methods, algorithms needed to make and test dynamic multiple regression models and time series (ARMA) models. In addition to teaching and learning methods based on the textbook, problem-based learning (PBL) and interactive engagement (IE) are used. Many internet data bases, EXCEL add-ins and EViews are used to enhance IE based learning. Selected SPSS or STATA examples are also provided. |
| BUS 340 | Management Science | FALL 2020 |
| In the first part of this computer-based course, students learn linear programming algorithms and how to apply them for resource allocation in production, investment selection, media selection, transportation planning, job assignments, financial planning, make or buy decision making and overtime planning contexts. In the second part of the course, students learn how to choose the best decision using expected monetary value (EMV), how to make optimum decision strategies under uncertainty by making decision trees, how to evaluate marketing research information, and how to apply project management (PERT) basic steps. Ultimately students are asked to conduct a month-long research and development project to define a real organizational decision strategy. |
| MAT 201 | Introduction to Statistics | FALL 2020 |
| This computer-based course presents the main concepts in Statistics: the concept of random variables, frequency, and probability distributions, variance and standard deviation, kurtosis and skewness, probability rules, Bayes theorem, and posterior probabilities. Important statistical methods like Contingency analysis, ANOVA, Correlation analysis and Regression Analysis are introduced and their algorithms are fully explained. The most important probability distributions are introduced: Binomial, Poisson, and Normal distribution, as well as the Chebyshev theorem for non-known distributions. Inferential statistics, sampling distributions, and confidence intervals are covered to introduce statistical model building and single linear regression. Active learning and algorithmic learning are stressed.
Emphasis is put both on algorithms –methods and assumptions for their applications. Excel is used while calculators with STAT buttons are not allowed. Ultimately students are required to make a month-long research project, select the theoretical concept they want to test, perform a literature review, find real data from Internet databases or make their surveys, apply methods they studied in the class, and compare theoretical results with their findings. Research is done and presented in groups, papers are Individual. Selected SPSS or Excel Data Analysis examples are also provided.
|
| BUS 306 | Quantitative Methods and Dynamic Forecasting | SPRING 2021 |
| In the first part of this course students learn concepts in inferential statistics, its main principles and algorithms. They learn how to apply sampling distributions in the case of business random variables, how to state and test business hypotheses about population mean or proportion differences, how to calculate ANOVA table components, and how to deploy estimation methods to provide information needed to solve real business problems. In the second part of the course, students learn advanced model building methods, algorithms needed to make and test dynamic multiple regression models and time series (ARMA) models. In addition to teaching and learning methods based on the textbook, problem-based learning (PBL) and interactive engagement (IE) are used. Many internet data bases, EXCEL add-ins and EViews are used to enhance IE based learning. Selected SPSS or STATA examples are also provided. |
| BUS 397 | Data Mining (Business Intelligence) | SPRING 2021 |
| This course introduces the cutting-edge computing methods for the analysis of business and marketing big data which help in inferring and validating patterns, structures and relationships in data, as a tool to support decisions at all levels of management. Students learn key descriptive, predictive, and prescriptive data mining methods with both supervised and non-supervised machine learning algorithms, which produce information for non-structured and semi structured decision making. While the course introduces a systems approach to business data processing, emphasis will be given to empirical applications using modern software tools such as Data Mining in Solver-Analytics More specifically, students will become familiar with and demonstrate proficiency in applications such as Cluster Analysis, Market Basket Analysis. Logistic Regression, Naïve Bayes Classification, Entropy Calculation, Classification Trees. Engagement-based learning is provided by using real world cases as well as computer based hands-on for real data analysis. Ultimately, working in teams, students will make the month long projects in applying Data Mining analytical techniques on the real world business problems, and will make suggestions for improvement which will be backed by the new information, gained from DM. Projects are presented in groups. Research papers, which are based on the projects, are individual. |
| MAT 201 | Introduction to Statistics | SPRING 2021 |
| This computer-based course presents the main concepts in Statistics: the concept of random variables, frequency, and probability distributions, variance and standard deviation, kurtosis and skewness, probability rules, Bayes theorem, and posterior probabilities. Important statistical methods like Contingency analysis, ANOVA, Correlation analysis and Regression Analysis are introduced and their algorithms are fully explained. The most important probability distributions are introduced: Binomial, Poisson, and Normal distribution, as well as the Chebyshev theorem for non-known distributions. Inferential statistics, sampling distributions, and confidence intervals are covered to introduce statistical model building and single linear regression. Active learning and algorithmic learning are stressed.
Emphasis is put both on algorithms –methods and assumptions for their applications. Excel is used while calculators with STAT buttons are not allowed. Ultimately students are required to make a month-long research project, select the theoretical concept they want to test, perform a literature review, find real data from Internet databases or make their surveys, apply methods they studied in the class, and compare theoretical results with their findings. Research is done and presented in groups, papers are Individual. Selected SPSS or Excel Data Analysis examples are also provided.
|
| BUS 340 | Management Science | SUMMER 2021 |
| In the first part of this computer-based course, students learn linear programming algorithms and how to apply them for resource allocation in production, investment selection, media selection, transportation planning, job assignments, financial planning, make or buy decision making and overtime planning contexts. In the second part of the course, students learn how to choose the best decision using expected monetary value (EMV), how to make optimum decision strategies under uncertainty by making decision trees, how to evaluate marketing research information, and how to apply project management (PERT) basic steps. Ultimately students are asked to conduct a month-long research and development project to define a real organizational decision strategy. |
| MAT 201 | Introduction to Statistics | SUMMER 2021 |
| This computer-based course presents the main concepts in Statistics: the concept of random variables, frequency, and probability distributions, variance and standard deviation, kurtosis and skewness, probability rules, Bayes theorem, and posterior probabilities. Important statistical methods like Contingency analysis, ANOVA, Correlation analysis and Regression Analysis are introduced and their algorithms are fully explained. The most important probability distributions are introduced: Binomial, Poisson, and Normal distribution, as well as the Chebyshev theorem for non-known distributions. Inferential statistics, sampling distributions, and confidence intervals are covered to introduce statistical model building and single linear regression. Active learning and algorithmic learning are stressed.
Emphasis is put both on algorithms –methods and assumptions for their applications. Excel is used while calculators with STAT buttons are not allowed. Ultimately students are required to make a month-long research project, select the theoretical concept they want to test, perform a literature review, find real data from Internet databases or make their surveys, apply methods they studied in the class, and compare theoretical results with their findings. Research is done and presented in groups, papers are Individual. Selected SPSS or Excel Data Analysis examples are also provided.
|
Journal Publications:
Book Chapter:
“Evaluating Realized Volatility Models with Higher Order Cumulants: HAR-RV Versus ARIMA-RV”. In H.Bilgin, M., Danis at al (eds). Eurasian Business Perspectives. Eurasian Studies in Business and Economics, Springer Nature-Switzerland. 2019. pp 315-336. DOI: 10.1007/978-3-030-11833-4
JOURNAL PUBLICATIONS:
"Testing GARCH and RV Exchange Rate Volatility Models using Hinich Tricorrelations," Oxford Journal on Business and Economics, Vol. 9, No. 2 (2014) ,136-151.
"A Cumulant-based stock market volatility modeling – Evidence from the international stock markets," Journal of Finance and Accountancy, Vol. 17, October 2014, 80-97.
http://www.aabri.com/manuscripts/141970.pdf
"Credit Spread Modeling: Macro-financial versus HOC approach," Economic Analysis, 2014, Vol. 47, No. 3-4, 53-68
https://ideas.repec.org/a/ibg/eajour/v47y2014i3-4p53-68.html
"Exchange Rate Volatility Forecasting Using Higher Order Cumulant Function." 2013. China-USA Business Review, ISSN 1537-1514, Vol. 12, No. 6, 533-542, ISSN 1573-1514
“A New Approach to Causality Testing”, Journal of Economic Analysis Vol. 44, no. 2011/1-2. EBSCO journal.
http://www.ien.bg.ac.rs/index.php/en/journal-economic-analysis
"Stock Market Volatility Forecasting Using Higher Order Statistics - Evidence from the Belgrade Stock Exchange." Statistical Review, Vol. LVIII, No 3-4, pp.19-30, 2009, ISSN 0039-0534.
http://scindeks.nb.rs/issue.aspx?issue=7757&lang=en
“Suppression of the regular Interference in the presence of Band-Limited White Noise, in: Torres: Signal Processing, Theories and Application.” IEEE Press, Elsevier Pub. Comp, Vol. 1, 593-597.
Working Papers:
” Inverse Time Series approach to Non Gaussian GARCH-ARMA FX Volatility Modeling “, Paper accepted for presentation at the 28th EBES Conference, London May 28-31, 2019.
"Micro and Macro Determinants of Stock Prices and their Volatility - The Case of International Oil Companies", Presented at 25th EBES Conference and published in the Proceedings, Berlin May 23-25, 2018.
“Does Multiplicative Seasonal Volatility Model Outperform Cascade Volatility Model? Evidence from Forex Market”. 22 EBES conference in May, Italy at the University “LA Sapienza”, 2017
"Testing GARCH and RV Exchange Rate Volatility Models using Hinich Tricorrelations-The case of Exchange Rate Volatility Forecasting," Proceedings of the CBEC conference, Cambridge 204, July 1-3, 2014.
"Entropy Loss in Non Gaussian High Frequency Measurement Systems," the 2014 International Conference on Systems and Informatics (ICSAI 2014) ,15-17 November 2014, Shanghai, China .
"Capturing stylized facts of Exchange Rate Volatility Using Higher Order Cumulant Function", Proceedings of the Cambridge Business & Economics Conference (CBEC), July 1-3. Cambridge, UK 2013.
http://www.gcbe.us/2013_CBEC/data/Sanja%20Dudukovic.pdf
“Exchange Rate Volatility Forecasting using Higher Order Cumulants and HF”, Proceedings of the Cambridge Business & Economics Conference (CBEC), June 27-29. Cambridge University, Cambridge, UK, 2012.
http://gcbe.us/2012_CBEC/data/Sanja%20Dudukovic.pdf
"Stock Market Volatility Forecasting - Evidence from the International Stock Markets.", Proceedings of the Cambridge Business & Economics Conference, Cambridge , UK, June 27-30, 2011.
http://www.gcbe.us/2011_CBEC/data/Sanja%20Dudukovic.pdf
”Answering the Skeptics: No, GARCH Exchange Rate Volatility Models Do Not Provide Accurate Forecasts”, Proceedings of the fourth EBES 2011 Conference (Eurasia Business and Economics Society) - Istanbul, June 1-3, 2011, Turkey.
"Spectral Analysis of GDP Shocks in US and BRIC Countries." Proceedings of the 4th Oxford Business and Economics Conference, Oxford, UK, June 28-30, 2010.
http://www.gcbe.us/2010_OBEC/data/Sanja%20Dudukovic.pdf
"Causality Testing Using Higher Order Cumulants." Proceedings of the 3rd Oxford Business and Economics Conference, Oxford, UK, June 23-26, 2009.
http://www.gcbe.us/2009_OBEC/data/Sanya%20Dudukovic.pdf
"Credit Spread Modeling Using Higher Order Cumulants." Proceedings of the Oxford Business and Economics Conference. ISBN: 978-0-9742114-7-3.
http://www.gcbe.us/2008_OBEC/data/Sanya%20Dudukovic%20.pdf
"Harmonic Analysis of the Real Business Cycles." Proceedings of the 2nd Oxford Conference on Business and Economics, Oxford, UK, June 24-27, 2007.
http://www.gcbe.us/2007_OBEC/data/Sanya%20Dudukovic.doc
"Causality between Interest Rate, Budget Deficit and Debt." Presented at the 6th International Conference on Business and Economics, Harvard University, USA, October 15-17, 2006.
"Forecasting Credit Spread Using Macro-Financial Variables." Proceedings of the 5th Global Conference on Business and Economics, Cambridge, UK, July 6-8, 2006. ISBN: 0-9742114-3-7.
“Micro-financial Determinants of Corporate Credit Spread," accepted for presentation at 5th Global Conference on Business & Economics Cambridge, UK, July 6-8, 2006.
“Dynamic Time Series Approach to Credit Spread Modeling," Proceedings of the Fourth Global Conference on Business & Economics, Jun, 26-28, 2005,Oxford, ISBN: 0-9742114-3-5.
“Feedback Between US Composite Index and Credit Spread,” Proceedings of 3rd International Finance Conference, IFC, 3-5 March 2005, Hammamet, Tunisia.
“Feedback between US Composite Index and Credit Spread.” Full paper accepted for the presentation on 3rd International Finance, Tunis, March 2005.
“Interest Rate Spread and GDP Growth –GARCH based Causality Test," Preoceedings of Global Conference on Business and Economics, July 7-9, 2004
"Extraction of the Band-Limited White Noise in the Presence of Colored Non Gaussian Noise," 3rd WSEAS International Conference on System Theory And Scientific Computation, poster session, Rhodos, Greece, July 28-30, 2003.
Awards and Honors:
2006: Franklin Faculty Excellence Award: Teaching
2011: Franklin Faculty Excellence Award: Teaching
2012: Franklin Faculty Excellence Award: Teaching
