Using GARCH models to predict prices and returns of stocks listed on the Iraq Stock Exchange
(An analytical study on a group of companies listed on the Iraq Stock Exchange)

Thesis Submitted to the Council of the College of Administration and Economics, Karbala University, in Fulfillment of the Requirements for Master’s degree in Finance and Banking

BY
Yusur Adnan Khalaf

Supervised by professor
Haider Abbas Aljanabi

Abstract
The study aims to use GARCH models in predicting stock price fluctuations and returns to assist in asset allocation and hedging, risk management, and portfolio optimization decisions. In addition to reducing errors in forecasting by calculating errors in advance forecasting and enhancing the accuracy of ongoing forecasts using the model.

This study was applied to a group of Iraqi companies and banks listed on the Iraqi Stock Exchange, which consists of 73 companies and banks. The study sample, which included 9 companies and banks, was selected for the period from January 2013 to December 2022, using several financial, statistical, and mathematical methods.

The study reached a number of conclusions, the most important of which is that GARCH models demonstrate their predictive ability through their reliance on a set of related matrices for the purpose of producing the best analytical results for the investor. As well as the ability to detect changes in the temporal structure of prices and returns, such as temporal fluctuations.

The study came out with a number of recommendations, the most important of which is: Before using GARCH models for forecasting, the historical data for the stocks in question must be analyzed, as the analysis must include a sufficient period of time to correctly estimate the parameters and determine the most appropriate model according to the nature of the financial data and the statistical characteristics of prices or returns.