Abstract
Cost overruns are common on long-term construction projects. This is mostly because of inaccurate early estimates and unexpected changes in the economy and finances. In Egypt, the costs of materials like steel, cement, bricks, sand, and aggregates make up a large part of the cost of building. These costs are greatly affected by the state of the economy and the financial markets. Even though the Construction Cost Index (CCI) is a widely used economic indicator around the world, Egypt has not yet made its own CCI official. This study creates a predictive model just for Egypt's construction industry that aims to predict a localized CCI to improve financial planning and lower risk. The framework uses two deep learning models, Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU), to make predictions about Egypt's CCI. The models include a wide range of macroeconomic, monetary, foreign exchange market, commodity/energy market and equity market indicators, as well as technical indicators. In Python, advanced statistical methods like correlation analysis, multicollinearity, and stepwise regression are used to make sure that the best features are chosen. The GRU is better at keeping things in balance because it wins on the calibration (Weighted Absolute Percentage Error (WAPE), Bias (mean error)), the absolute error metrics (Mean Absolute Error, Mean Absolute Percentage Error, Symmetric Mean Absolute Percentage Error, and median error), while LSTM is better at squared-loss/association and turning points (Root Mean Squared Error, Mean Squared Error, Coefficient of determination, Directional Accuracy) because it has a slightly tighter variance fit and sign tracking. There is a permutation feature importance analysis for six features in both the GRU Model and the LSTM Model that shows that oil is the most important thing that affects the construction cost index (CCI). The study shows that deep learning models can accurately predict economic indicators. This gives Egypt's construction industry a useful, data-driven tool for estimating costs ahead of time. They make a big difference in Egypt's construction industry and meet the need for localized forecasting models in markets.