COURSE UNIT TITLE

: ESTIMATION AND FORECASTING BY ECONOMETRIC MODELS

Description of Individual Course Units

Course Unit Code Course Unit Title Type Of Course D U L ECTS
EKO 6056 ESTIMATION AND FORECASTING BY ECONOMETRIC MODELS ELECTIVE 3 0 0 6

Offered By

Econometrics

Level of Course Unit

Third Cycle Programmes (Doctorate Degree)

Course Coordinator

ASSOCIATE PROFESSOR MURAT TANIK

Offered to

Econometrics

Course Objective

The aim of this course is to teach forecasting methods, the algorithms for combining forecasts to improve forecast accuracy, and show the entire process for the principles of forecasting by using R program to students who will be able to not only have theoriccal knowledge, but also have pratical skill by anlayzing the time series data at the end of the couse.

Learning Outcomes of the Course Unit

1   1. To be able to indetify the characteristics of time series data, and to adjust the data before the analyzes,
2   2. To be able to decide the forecasting mehods based on data,
3   3. To be able to introduce hybrid forecasting methods instead of classical time series forecasting models,
4   4. To be able to learn the importance of using recursive and rolling window appraches in terms of forecasting performances,
5   5. To be able to combine different models or even algorithms to improve forecast accuracy,
6   6. To be able to understand the advantages of using adaptive algrithms in big data forecasting.

Mode of Delivery

Face -to- Face

Prerequisites and Co-requisites

None

Recomended Optional Programme Components

None

Course Contents

Week Subject Description
1 1. Forecasting Terminology and notations
2 2. Decription of characteristic features of a time series by using plots (time plot, seasonal plots, polar seasonal plots, decomposition plots, autocorrealtion plot)
3 3. Forecasting performances measures: Advantages and disadavantages against each other (MAE, RMSE, MASE, MDB, Theil U and others)
4 4. Widely-used forecasting models in literature (ARMA, sARIMA, Holt-Winters, TBATS. ETS. VAR, VECM, ARDL, ANN)
5 5. Slection of the most proper forecasting model based on the features of data.
6 6. Out-of-sample forecasting performances of forecasting models (application with R)
7 7. Forecasting performances of hybrid methods (ARIMA-ANN, HW-ANN, ETS-ANN and others)
8 8. Finding the optimal training size for the best forecasting performance (application of recursive window and rolling window)
9 9. Comparison tests for forecasting models (Single and multiple tests)
10 10. Combining algorithms to improve forecast accuracy (Mean, Winsorized mean, Median, Trimmed mean, Bates/Granger (1969) algorithm, Newbold/Granger (1974) algorithm and inverse rank)
11 11. Other algorithms to combine forecasts (methods based on variance-covariance, OLS regression, LAD regression and neural networks)
12 12. Finding a optimal number of combined models (application with R)
13 13. Adaptive algortihms for big data forecasting (Empirical Mode Decomposition and Wavelet Decomposition)
14 14. R application by using real-word datasets

Recomended or Required Reading

Hyndman, R.J., & Athanasopoulos, G. (2018) Forecasting: principles and practice, 2nd edition, OTexts: Melbourne, Australia.
Chatfield, C. (2000). Time-series forecasting. CRC press.
Chatfield, C., & Xing, H. (2019). The analysis of time series: an introduction with R. CRC press.
Brockwell, P. J., Davis, R. A., & Calder, M. V. (2002). Introduction to time series and forecasting (Vol. 2, pp. 3118-3121). New York: springer.
Ramasubramanian, K., & Singh, A. (2018). Machine Learning Using R: With Time Series and Industry-Based Use Cases in R. Apress.

Planned Learning Activities and Teaching Methods

Lecture Method, Question-Answer Method, Method Discussion and Problem Solving Method

Assessment Methods

SORTING NUMBER SHORT CODE LONG CODE FORMULA
1 MTE MIDTERM EXAM
2 STT TERM WORK (SEMESTER)
3 FIN FINAL EXAM
4 FCG FINAL COURSE GRADE MTE * 0.30 + STT * 0.30 + FIN* 0.40
5 RST RESIT
6 FCGR FINAL COURSE GRADE (RESIT) MTE * 0.30 + STT * 0.30 + RST* 0.40


Further Notes About Assessment Methods

None

Assessment Criteria

To be announced.

Language of Instruction

Turkish

Course Policies and Rules

To be announced.

Contact Details for the Lecturer(s)

emrahgulay2011@gmail.com

Office Hours

To be announced.

Work Placement(s)

None

Workload Calculation

Activities Number Time (hours) Total Work Load (hours)
Lectures 13 3 39
Preparations before/after weekly lectures 13 3 39
Preparation for midterm exam 1 40 40
Preparation for final exam 1 40 40
Midterm 1 3 3
Final 1 3 3
TOTAL WORKLOAD (hours) 164

Contribution of Learning Outcomes to Programme Outcomes

PO/LOPO.1PO.2PO.3PO.4PO.5PO.6PO.7PO.8PO.9
LO.1111333333
LO.2122222222
LO.3222222222
LO.4111111111
LO.5333322222
LO.6222222222