This study investigate the use of Bayesian network models in investment decisions under Saudia Arabia dynamic Market conditions. The objective is to identify the interrelated association among risk factors, asset performance and their returns, which capable to investors to make choices over changing environments. This study used a Bayesian network model which is based on three main factors expected return, risk level and investor confidence and investment decisions using dataset of 2011-2022. We also consider the political instability and interest factors that are linked with investor decision nodes, assets classes and strategies. These factor allows the simulations and their impact on potential assets return and risk profiles. The findings suggest that there is increase in political instability which negatively influence the investor confidence, in results it leads to decline in average return in emerging markets. The results reveals that high volatility in assets price, leads to portfolio composition volatility that required adjustment for risk tolerance levels and favor diversification into low volatile assets. Finally, interest rates found to be led to high historical expected return for fixed income securities in our portfolio. The findings highlights the importance of Bayesian network models in recognizing the key risk and return trade-off,  adapting dynamics in market changes, evaluate and improve the risk management for investor strategies. This study contributes in the investor confidence for making investment decisions, based on Bayesian network models in complex financial scenarios. The investors can optimize their portfolios and take actions in dynamic market uncertainties.