A novel MPK optimization framework for financial data analysis incorporating complexity and uncertainty management

Abstract

In a competitive environment, the ability to scale quickly and successfully is a critical need. This research proposes a new framework using multi-objective complexity prediction model (MPK) for financial data analysis, including complexity and uncertainty management. This model integrates input, uncertainty, and output optimization functions (OOFs) (input optimization function (IOF), uncertainty optimization function (UOF), and OOF) to predict complex output values under dynamic business conditions. Model evaluation is carried out using performance metrics, namely mean squared error (MSE), root mean squared error (RMSE), mean absolute error (MAE), and R² score. The evaluation results show that this model has an MSE value of 20.112, an RMSE of 2.267, and an MAE of 2.351, reflecting a low prediction error rate and high accuracy. In addition, the R² value of 0.884259 indicates that this model is able to explain around 88.4% of the variability in the output data, indicating its ability to capture complex data patterns. Thus, the proposed MPK model is effective in predicting output values in complex business scenarios and can be applied for strategic decision-making under conditions of uncertainty.