Thesis submitted
To the Council of the College of Management and Economics at the University of Karbala
It is part of Neil doctorate degree requirements in the philosophy of Banking and Finance

Presented By

Hussein Hadi Abdul Amir

The supervision By
Prof. DrAli Ahmed Faris

Abstract

This study aimed to use Bayesian estimation to predict stock returns under multiple loss functions (squared loss function and absolute loss function). The study was conducted in the Iraq Stock Exchange on a sample of (42) companies listed in the market, using the monthly data of the stock returns of these companies for the period from 2015 to 2022.

The Bayesian model was used to determine the relative probability of stock performance in the future using historical data in the hope of reducing the percentage of error in making the buying or selling decision by investors in the financial market.

The results showed that the use of Bayesian estimation can help improve the prediction accuracy of returns in the short and medium term, which reduces the relative error by up to 30% compared to other models.

The study also concluded that the use of Bayesian estimation in predicting stock returns under multiple loss functions can improve the prediction accuracy and reduce the relative error rate, and then these results can be used to improve the investment decision and achieve the highest possible levels of returns.

The study found the possibility of using Bayesian estimation models in light of multiple loss functions to improve the predictive ability of investors in the Iraq Stock Exchange. Hence, the study recommended (analysts and investors) to use these models to obtain the best predictive value of returns.

Thesis submitted
To the Council of the College of Management and Economics at the University of Karbala
It is part of Neil doctorate degree requirements in the philosophy of Banking and Finance

Presented By

Hussein Hadi Abdul Amir

The supervision By
Prof. DrAli Ahmed Faris

Abstract

This study aimed to use Bayesian estimation to predict stock returns under multiple loss functions (squared loss function and absolute loss function). The study was conducted in the Iraq Stock Exchange on a sample of (42) companies listed in the market, using the monthly data of the stock returns of these companies for the period from 2015 to 2022.

The Bayesian model was used to determine the relative probability of stock performance in the future using historical data in the hope of reducing the percentage of error in making the buying or selling decision by investors in the financial market.

The results showed that the use of Bayesian estimation can help improve the prediction accuracy of returns in the short and medium term, which reduces the relative error by up to 30% compared to other models.

The study also concluded that the use of Bayesian estimation in predicting stock returns under multiple loss functions can improve the prediction accuracy and reduce the relative error rate, and then these results can be used to improve the investment decision and achieve the highest possible levels of returns.

The study found the possibility of using Bayesian estimation models in light of multiple loss functions to improve the predictive ability of investors in the Iraq Stock Exchange. Hence, the study recommended (analysts and investors) to use these models to obtain the best predictive value of returns.