ADS Capstone Chronicles Revised

making the model unsuitable for traders to rely on for daily or weekly decision-making. Additional factors need to be considered to improve the model ’ s generalizability. The model demonstrates the feasibility of forecasting but

needs to be more reliable for making sound trading decisions. However, it can help understand specific patterns but should not be used as the sole basis for trading actions.

Figure 10 Forecasted Values vs Actual Values

cryptocurrency price prediction, instilling confidence in the audience.

8 Discussion One of the key findings from this study was the significant performance of the Ridge Regression model. Despite the strong performance of multiple models, the Ridge Regression model, a linear regression model that uses L2 regularization to prevent overfitting, stood out. It demonstrated outstanding performance on the training and test sets, surpassing initial expectations. Its consistency and reliability across all metrics make it a strong candidate for implementation in a Streamlit application for

8.0.1 Model Selection The decision not to choose other top-performing models was driven by their less consistent performance across the train and test evaluations. The robustness of the Ridge Regression model, despite its slight tendency towards overfitting, was a critical factor in the decision. This robustness, which can be managed

87

Made with FlippingBook - Online Brochure Maker