Master 2019/2020

Time Series Analysis-1

Type: Elective course (Applied Economics)

Area of studies: Economics

Delivered by: Department of Applied Economics

Where: Faculty of Economic Sciences

When: 1 year, 3 module

Mode of studies: offline

Instructors: Grigory Kantorovich, Dmitry Malakhov

Master’s programme: Applied Economics

Language: English

ECTS credits: 3

Contact hours: 40

Course Syllabus

Abstract

This course provides tools for quantitative analysis of economic time series, for example, macroeconomic and financial series. The course is an introduction to univariate statistical methods of analysis and forecasting of economic and financial time-series. ARIMA(p,d,q,) model, stationarity concept, Box-Jenkins approach and different forecasting issues are discussed in this course. Models and concepts from this course can be used in stochastic processes, mathematical models in economics, optimal control problems, statistical forecasting, financial mathematics, decision making under uncertainty. The stress in the course is made on the sense of facts and methods of time series analysis. Conclusions and proofs are given for some basic formulas and models; this enables the students to understand the principles of economic theory. The main stress is made on the economic interpretation and applications of considered economic models.

Learning Objectives

• providing tools for quantitative analysis of economic time series, for example, macroeconomic and financial series

Expected Learning Outcomes

• get acquainted with the main concepts of time series theory and methods of analysis
• know how to use them in examining economic and financial processes
• understand methods, ideas, results and conclusions that can be met in the majority of books and articles on economics and finance
• should master traditional methods of time series analysis, intended mainly for working with time series data
• understand the differences between cross-sections and time series
• understand specific economic problems, which occur while working with data of these types
• get knowledge about how to analyze economic time series, what potential problems and pitfalls can exist in such cases
• get knowledge about how to compare different statistical models using in-sample and out-of-sample statistical measures of accuracy

Course Contents

• Time series as a discrete stochastic process. Stationarity of stochastic processes.
• Autoregression-moving average model ARMA(p,q). Autocorrelation and partial autocorrelation functions.
• Estimation of coefficients in ARMA(p,q) processes. Information criteria.
• Forecasting using Box-Jenkins approach.
• Nonstationary time series. Box-Jenkins approach for determination of a degree of integration of time series.

Assessment Elements

• Home assignment
  No late submissions of the home assignment are allowed.
• Final test
  The examination is conducted in writing (tasks / open-ended questions) using synchronized proctoring. At the beginning of the exam, students are sent an exam file and assignments to the group mail and LMS. At the end students should send readable photos/scans of their exams to the mentioned in the exam file email. Students have 7 minutes after exam in order to send photos. Proctoring during the exam takes place on the Examus platform (https://hse.student.examus.net). In order, the exam to be considered by grader, the student must connect to this platform. You must connect to the platform 15 minutes before the exam. On the Examus platform, system testing is available. Student computer must meet the following requirements (see attachement). To participate in the exam, the student must: go to the proctoring platform in advance, conduct a system test, turn on the camera and microphone, and verify identity. During the exam, students are prohibited from: communicating (on social networks, with people in the room), using of electronic or written notes or any other electronic devices. You can use only the built-in OS simple calculator. A short-term communication disruption during the exam is considered interruption of connection up to 5 minutes. A long-term communication disruption during an exam is considered interruption of connection of 5 minutes or more. In case of a long-term communication disruption, the student cannot continue the exam. The exam retake is similar to the original. If a proctor notices student’s suspicious actions during the exam or unusually high results are identified, video interviews will be conducted.

Interim Assessment

• Interim assessment (3 module)
  0.7 * Final test + 0.3 * Home assignment

Bibliography

Recommended Core Bibliography

• Applied econometric time series, Enders, W., 2004
• Канторович Г.Г. (2002). Лекции: Анализ Временных Рядов. Higher School of Economics Economic Journal Экономический Журнал Высшей Школы Экономики, (1), 85. Retrieved from http://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edsrep&AN=edsrep.a.scn.025886.16537823
• Канторович Г.Г. (2002). Лекции: Анализ временных рядов. Retrieved from http://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edsbas&AN=edsbas.69D6F004
• Канторович Г.Г. (2002). Лекции: Анализ временных рядов. Экономический Журнал Высшей Школы Экономики, (1). Retrieved from http://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edsclk&AN=edsclk.16537823
• Канторович, Г. (2002). Лекции: Анализ Временных Рядов. Экономический Журнал Высшей Школы Экономики, (3). Retrieved from http://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edsclk&AN=edsclk.15693389
• Путеводитель по современной эконометрике : учеб.- метод. пособие для вузов, Вербик, М., 2008

Recommended Additional Bibliography

• Hamilton, J. D. . (DE-588)122825950, (DE-576)271889950. (1994). Time series analysis / James D. Hamilton. Princeton, NJ: Princeton Univ. Press. Retrieved from http://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edswao&AN=edswao.038453134
• Time series analysis, Hamilton, J. D., 1994