Figure 10
A scatter plot displays the residuals of a random forest model, with the x-axis representing observations and the y-axis representing residuals. The plot shows dozens of data points scattered around the zero line, indicating the difference between actual and predicted values. There are noticeable clusters of positive and negative residuals, particularly during periods of extreme price fluctuations. The residuals are not evenly distributed, suggesting systematic prediction errors and large deviations during price spikes. The presence of structured residual patterns indicates that certain temporal dynamics and external factors influencing price movements are not fully captured by the model. This pattern confirms that the model struggles to accurately predict high-value observations and price spikes. All values are approximated.Residual plot for random forest model showing systematic prediction errors and large deviations during price spikes. (Source: Authors’ own work)