Workshop Objectives

The announced training workshop, Applied Multivariate Time Series Analysis: Focusing on the VAR Family, will provide participants with a comprehensive understanding of vector autoregressive (VAR) models and their applications in economic and financial data analysis. By focusing on VAR, VECM (vector error correction models), Granger causality, and SVAR (structural VAR), the workshop will equip participants with the technical skills needed for effective modeling, analysis, and interpretation of interdependent time series data. Participants will gain both theoretical knowledge and practical experience by utilizing software for model estimation and testing. 

Outcomes
Upon completion, participants will:

• Have a clear understanding of the VAR model family, including VAR, VECM, Granger Causality, and SVAR, and their distinct applications.

• Develop skills to analyze complex interdependencies among multiple time series.

• Learn to identify and apply the appropriate model based on data structure and research objectives.

• Gain experience using statistical software to estimate models, conduct diagnostics, and interpret results.

Prerequisites
Participants should have:

• Understanding of Econometrics and Time series analysis.

• Familiarity with statistical software EViews.

• Elementary Matrix Algebra (Multiplication and Inverse)

• Ideally candidate bring own laptop

Classes Schedule:

Commencement of the course is Tuesday, December 10, 2024.

Session 1  Tuesday, December 10, 2024.   6PM to 8:30 PM

Session 2 Thursday, December 12, 2024 6PM to 8:30 PM

Session 3 Tuesday, December 17, 2024. 6PM to 8:30 PM

Session 4 Thursday, December 19, 2024 6PM to 8:30 PM

Course Content 

• Introduction Time Series Data

• Properties of Time Series Data and Stationarity, Unit Root Test

• Unit root with Structural Breaks

• Vector Autoregressive (VAR) Model 

  • Introduction

  • Lag Length Selection

  • Diagnostic Test/ Model Checking

  • Impulse Response

  • Variance Decomposition

• Granger Causality 

• Cointegration and Error Correction Mechanism 

• Vector Error Correction Model (VECM)

• Linear Restriction in VAR

• SVAR/ VAR with contemporaneous Relation

  • Identification Problem

    • Recursive identification: Sims (1992)

    • Non Recursive: Blanchard and Perotti (2002)

  • Short Run Restrictions

  • Long Run Restrictions