Applied Time Series Analysis-Advance your Knowledge Be a Data Scientist
Applied Time Series Analysis-Advance your Knowledge Be a Data Scientist
Applied Time Series Analysis
Advance Your Knowledge, Be a Data Scientist
Course Overview
This comprehensive short course is designed to equip participants with the essential tools and techniques for analyzing time series data. Participants will gain practical experience in handling real-world data, understanding time series properties, and mastering advanced models to make data-driven decisions. This course focuses on hands-on learning using EViews software, ensuring that concepts are both theoretical and application-driven.
What You Will Learn:
Fundamentals of data types, regression analysis, and model diagnostics
Stationarity concepts, unit root tests, and transformations
ARIMA, VAR, and ARDL models for time series analysis
Cointegration, Vector Error Correction Models (VECMs), and long-run relationships
Structural VAR models and contemporary forecasting techniques
Real-world forecasting applications and model evaluation
Course Modules:
Module 1: Data Types, Regression Analysis, and Diagnostics
Module 2: Stationarity and Unit Root Tests
Module 3: ARIMA Models and Forecasting
Module 4: Vector Autoregressive (VAR) Models
Module 5: ARDL Models and Cointegration
Module 6: Vector Error Correction Models (VECMs)
Module 7: Structural VAR Models
Module 8: Forecasting Techniques and Accuracy Evaluation
Software Used: EViews for hands-on exercises and implementation
Who Should Attend?
Graduate students, researchers, and professionals seeking expertise in time series analysis
Data scientists and analysts aiming to enhance forecasting and modeling skills
Economists, statisticians, and policy analysts working with dynamic data
Classes Schedule:
The course will commence on Saturday, January 11, 2025. Classes are scheduled for every Saturday from 10 AM to 1 PM
Duration: 8 Lectures
Location: Applied Economics Research Centre(AERC), University of Karachi.
Instructor: Dr. Syed Ammad Ali
Course Content
Module 1: Introduction to Data and Regression Analysis (Lecture 1-2)
1. Understanding Data Types and Structures
o Overview of data types: Time series, cross-sectional, and panel data
o Characteristics of time series data (trends, seasonality, cyclicity)
o Data visualization for time series analysis
2. Fundamentals of Regression Analysis
o Introduction to linear regression
o Assumptions of Ordinary Least Squares (OLS)
o Multivariate regression and interpreting coefficients
o Role of regression in time series analysis
3. Diagnostics for Regression Models
o Residual analysis: Checking for homoscedasticity and normality
o Autocorrelation and multicollinearity: Identification and remedies
o Goodness-of-fit measures: R-squared, adjusted R-squared, and AIC/BIC
o Importance of model diagnostics in forecasting
Module 2: Time Series Data and Stationarity (Lecture 3)
1. Introduction to Time Series Data
o Understanding lags, differences, and transformations
o Visualizing time series data using plots
2. Stationarity in Time Series
o Concept of stationarity and non-stationarity
o Testing for stationarity: ACF, PACF, and unit root tests (Augmented Dickey-Fuller Test, PP-test etc)
o Addressing non-stationarity with differencing and transformations
o Unit root with Structural Breaks
3. Practical Session in EViews
o Importing and visualizing time series data
o Performing unit root tests in EViews
Module 3: Autoregressive and ARIMA Models (Lecture 3)
1. Autoregressive Models
o Autoregressive (AR) and Moving Average (MA) models
o Combining AR and MA: ARIMA models
o Identifying ARIMA models: ACF, PACF, and order selection
2. Model Estimation and Diagnostics
o Estimation of ARIMA models in EViews
o Checking model assumptions and diagnostics
o Model validation using residual analysis
3. Practical Exercises in EViews
o Building and validating ARIMA models
o Forecasting using ARIMA models
Module 4: Vector Autoregressive (VAR) Models (Lecture 4)
1. Introduction to VAR Models
o Basics of VAR models and their applications
o Lag length selection and stability conditions
2. Outcomes of VAR Models
o Impulse response functions (IRFs)
o Variance decomposition
3. Granger Causality
o Introduction and Application
4. Practical Implementation in EViews
o Estimating and interpreting VAR models
o Analyzing IRFs and variance decomposition results
Module 5: ARDL Models and Cointegration (Lecture 5)
1. Autoregressive Distributed Lag (ARDL) Models
o Introduction to ARDL models and their advantages
o Application of ARDL models in time series data
2. Cointegration and Long-Run Relationships
o Concept of cointegration
o Bounds testing approach for ARDL models
3. Practical Exercises in EViews
o Estimating ARDL models
o Testing for cointegration and interpreting results
Module 6: Vector Error Correction Models (VECMs) (Lecture 6)
1. Introduction to VECMs
o Relationship between VAR and VECM
o Estimation of VECMs for cointegrated systems
2. Diagnostics for VECMs
o Residual analysis and model stability
o Impulse response functions in VECMs
3. Practical Session in EViews
o Estimating and interpreting VECMs
Module 7: Forecasting Techniques (Lecture 7)
1. Structural Vector Autoregressive Models(SVAR)
o Linear Restriction in VAR
o VAR with contemporaneous relationship
o Identification Problem
o Recursive identification: Sims (1992)
o Non Recursive: Blanchard and Perotti (2002)
o Short Run Restrictions
o Long Run Restrictions
2. Practical Forecasting Exercises in EViews
o Estimating and interpreting SVAR
Module 8: Forecasting Techniques (Lecture 8)
1. Forecasting in Time Series Models
o Importance of forecasting in decision-making
o Forecasting using ARIMA, VAR, ARDL, and VECM models
2. Model Based Forecasting
o Building Models and System based Forecasting
3. Forecast Evaluation
o Forecast accuracy measures: RMSE, MAE, MAPE
o Comparing forecast performance of different models
4. Practical Forecasting Exercises in EViews
o Generating and evaluating forecasts
For AERC Students and Alumni …............................……………………PKR. 6,000
Students other Departments/Universities ………………………….. PKR. 8,000
Professionals …………………………………………........................................PKR. 10,000
Please deposit your fees (online/Cash) into the designated account and retain a copy of your payment receipt .
A/C Title: ENDOWMENT FUND ACCOUNT AERC
NBP- UNIVERSITY OF KARACHI BRANCH
PK69 NBPA 0071 0041 000 65972