Time Series Forecasting and Data Quality

Online course: October 10 – 24, 2018IWorkshop: November 6 – 8, 2018ILjubljana, SloveniaIData and Analysis for Designing Policies, Budget Preparation and Execution, Central Banking

This blended learning was a continuation of the workshop on the topics on forecasting methodologies, data collection and quality, that the CEF organized in 2017. The aim of this learning initiative was to strengthen the knowledge and the skills of central bank and finance officials from CEF constituency and beyond, in time series analysis and forecasting.

What did you learn?

The main goal of this blended learning initiative was to bring together finance officials to refresh their knowledge in time series analysis as well as to enhance their forecasting techniques of selected macroeconomic variables (e.g. GDP, inflation, exports), in EViews.

This learning initiative was delivered as blended learning course, consisted of both e-learning phase and a workshop.

The learning initiative was divided into:

1.         E-learning phase: The online learning phase started on October 10, 2018 and finished on October 24, 2018. It served as preparation for the face-to-face part of the course and will be delivered through the CEF Online Learning Campus: https://www.cef-see.org/olc/.

However, participants that could not attend the workshop in Ljubljana,  participated only in the e-learning phase.

The envisaged workload for participants was about 3 hours per week, consisting of: webinar session, reading materials, quiz, assignment and an opportunity to connect with other course fellows to debate ideas and get help mastering addressed subject matters.

2.         Face-to-face workshop: at CEF in Ljubljana from November 6 to November 8, 2018.

The workshop provided lectures based on practical cases and methodological refresher, accompanied by practical sessions where participants will work on exercises with real data, using the EViews software package. With support from our IMF resident advisor on government finance statistics, we also addressed data quality. Participants had the opportunity to learn through practical examples how to do quality and consistency check within their data.

How did you benefit?

Upon completion of this blended learning initiative, the participants were be able to:

  • Elaborate the six question of forecasting and overview of the forecasting techniques
  • Explain the patterns of time series data (random, trend, seasonal, cyclical) and the decomposition of time series data
  • Understand Data quality issues and their importance for forecasting, as well as learn how to assess data quality
  • Understand the importance of macroeconomic forecasts from the economic policy perspective with an insight on past failures and biasedness in forecasting
  • Receive insights into ARMA models and ARDL models for forecasting and nowcasting
  • Deal with many variables and frequency mismatch in data
  • Learn how to use of time-series models (VAR) for forecasting with short time-series
  • Apply single equation, stationary VARs and BVARs (with macroeconomic data) on selected country variables in EViews

Who attended the learning initiative?

The workshop was designed for staff working in:

  • central banks - research and analyses / forecasting departments
  • ministries of finance – analysts engaged in producing macroeconomic forecasts
  • national statistical offices and statistical departments of central banks;
  • fiscal councils, non-government institutes, as well as research and academic institutions producing macroeconomic forecasts.

The workshop was delivered in English and highly interactive, where participants were encouraged to actively participate and share their experience and challenges on the covered topics.

Practical information

No fee was charged, and lunches and refreshments was provided for all selected participants. 

Applications needed to be submitt no later than October 2 via online application form.

Candidates will need to be approved by the CEF; selected candidates will be notified by October 5.


Laurent Ferrara, PhD - Banque de France

Laurent Ferrara is Head of the International Macroeconomics Division at the Banque de France in Paris, in charge of the outlook and macroeconomic forecasting for advanced economies, as well as global policy issues such as exchange rates, commodities or global imbalances. Main tasks of this division of around 20 people are policy briefing, preparation of international meetings (ECB, IMF, OECD, G20) and economic research. He is also involved in academics and has been appointed Adjunct Professor of Economics at the University of Paris Nanterre in September 2011.

Laurent Ferrara is Director of the International Institute of Forecasters, an international association aiming at bridging the gap between theory and applications in forecasting, through the organization of workshops and conferences and the publication of an academic journal, the International Journal of Forecasting. He is also an associate editor of this journal.

Dr. Ferrara holds a PhD in Applied Mathematics from the University of Paris North (2001) and a Research Habilitation in Economics from the University of Paris 1 – Panthéon – Sorbonne (2007). His academic research mainly focuses on macroeconomic forecasting, international economics, econometric methods, non-linear modelling and business cycle analysis. He published more than 50 papers in international and national academic journals, chapters in books, as well as book on time series analysis and forecasting.

Klodiana Istrefi, PhD - Banque de France 

Klodiana Istrefi is an Economist at the Monetary Policy Research Division at the Banque de France since 2014. She holds a PhD in Economics from Goethe University Frankfurt (2016). Previously she has worked as a Senior Economist at the Bank of Albania and as a Researcher at the Centre of Excellence SAFE in Frankfurt. Her current work focuses on studying uncertainty as perceived by market participants with regard to monetary policy. Her primary fields of interest are Monetary Economics, International Economics, Applied Macroeconomics and Macromodeling.

Ana Grdović Gnip, PhD - University of Primorska

Ana Grdović Gnip is an assistant professor at the University of Primorska, Faculty of Mathematics, Natural Sciences and Information Technologies. She holds a PhD in Economics from the University of Ljubljana, Faculty of Economics (2014). She participated as researcher in several scientific projects and as lecturer in different teaching programmes. Currently, she is also the academic coordinator of the Jean Monnet Module course “EU Economic Trends” financed by the European Commission. Her current research focuses on the efficiency of fiscal rules and fiscal consolidation. Her primary fields of interest are Fiscal studies, Applied Macroeconomics and Macromodeling.

Deon Florian Tanzer - International Monetary Fund

Deon Florian Tanzer works for the IMF and has been based at the CEF since September 2016. He is a technical advisor for government finance statistics (GFS) in South East Europe. He advises compilers in various national institutions on how to develop GFS data and data reporting, both for the IMF and the EU in compliance with the fiscal reporting requirements (the so-called Maastricht deficit and debt and the Excessive Deficit Procedure). Deon has gained extensive experience in GFS at the IMF and Statistics Netherlands. His work is funded by the Swiss Government.

The development of this learning initative is supported by CEF programme advisor Mr. Janez Šušteršič. He was director of the Institute of Macroeconomic Analyses and Development (IMAD) from 2000 to 2007. In that capacity, he led on development of forecasting approach from a purely judgmental one to incorporating advanced economic modeling methods. The Institute under his leadership developed tools for assessing impact of policy measures and advanced economic analysis. From 2014 to 2016, Mr. Šušteršič was a key expert on a technical assistance project in Kosovo, leading a team of experts that helped develop forecasting models and techniques used by their Ministry of Finance. From his experience, he has broad understanding of methodological, institutional and data quality issues involved in forecasting and practical solutions that may be used in imperfect environments. Currently, Mr. Šušteršič is a partner at Re-forma, ltd., working as international consultant and head of research group, and a Full Professor of Economics.


This learning initiative was supported by:

Bank of Slovenia Banque de France International Monetary Fund
Slovenia's Development Cooperation