Abstract

The construction industry is one of the main contributors to the social and economic development of any country, and the growth of the industry is highly dependent on the financial success of construction companies. Contractors are usually impacted financially by external factors such as the fluctuating material prices and challenging macroeconomic conditions, making it difficult for the company’s executives to make strategic decisions. The goal of this research is to assess the financial performance and valuate different construction contractors in Egypt by predicting their stock price in the future. This is achieved by developing three artificial intelligence-based models incorporating Long-Short Term Memory (LSTM). The first model would include the macroeconomic indicators with the stock price of each company to investigate how macroeconomic factors affects each company’s stock movement. The second model includes another external factor which is the competitors’ stock price as input variables for the model developed to examine how the market competition could affect each company’s stock price The third model considers internal financial metrics only to determine how the financial health of each company impacts its stock price regardless of the external influence. To develop these models, financial statements of 16 publicly listed contractors in Egypt were gathered to obtain their financial position in the last decade. The input parameters including external factors such as macroeconomic indicators and material prices, and internal factors such as revenues and working capital were used in the model. Correlation analysis was then performed between these identified variables and the stock price to remove any variable with weak correlation. The data was then split into training and testing, and LSTM was used in the developed models to analyse and predict the stock price. The first model provided mean absolute percentage error (MAPE) ranging between 20.62% to 72.8% for the tested companies. The second model provided improved results with reasonable (MAPE) ranging from 19.83% to 27.96%. The final model provided the best performance with MAPE ranging from 12.99% to 19.12%. Therefore, this research provides a comparison of the predictive performance of LSTM models based on macroeconomics indicators, competitor performance, and internal financial metrics specific for each company. This provides decision makers in the construction industry with practical tools to anticipate financial challenges and make strategic decisions to optimize the cash flow and improve their profitability by anticipating revenue fluctuations according to the provided variables. On the other hand, it could be used as a reliable framework for external investors to assess the performance of construction contractors, allowing them to make more informed investment decisions in challenging and emerging markets such as in Egypt.

School

School of Sciences and Engineering

Department

Construction Engineering Department

Degree Name

MS in Construction Engineering

Graduation Date

Fall 2-15-2026

Submission Date

1-26-2026

First Advisor

Dr. Ossama Hosny

Second Advisor

Dr. Ibrahim Abotaleb

Committee Member 1

Dr. Khaled Nassar

Committee Member 2

Dr. Mohamed Mahdy

Committee Member 3

Dr. Ahmed El Gendy

Extent

107 p.

Document Type

Master's Thesis

Institutional Review Board (IRB) Approval

Not necessary for this item

Disclosure of AI Use

Code/algorithm generation and/or validation

Download

Available for download on Tuesday, January 26, 2027

Share

COinS