Analyzing Chinese Banking Performance with a Trigonometric Envelopment Analysis for Ideal Solutions Model

Abstract

Non-parametric Data Envelopment Analysis (DEA) is susceptible to the curse of dimensionality, a challenge that can be mitigated through the use of the Multi-Criteria Decision-making (MCDM) method. Conversely, DEA can overcome the limitations of the MCDM method by defining the weights of the Decision-Making Unit to calculate the data envelop. This study addresses this issue by introducing a novel model, the Trigonometric Envelopment Analysis for Ideal Solutions (TEA-IS). TEA-IS combines DEA and the Technique for Order Performance by Similarity to Ideal Solution (TOPSIS) approaches. The proposed method is employed to assess the efficiency and performance of 367 Chinese banks over a 19-year period using various financial variables. The TEA-IS model leverages machine learning techniques to predict positive or negative outcomes for Chinese banks, taking into account various influencing factors. Our results indicate that TEA-IS scores demonstrate superior discriminatory power and reliability compared to non-parametric and MCDM methods. Furthermore, our findings reveal the presence of synergy among Chinese banks and illustrate a pattern of volatility in the Chinese banking industry’s performance. Notably, performance improved from 2000 to 2005, declined during the period from 2006 to 2013, and subsequently experienced a recovery until 2018. The majority of Chinese banks in the sample are categorized as medium performers with lower synergy levels. Additionally, the study underscores the positive impact of bank listing and age on bank performance, suggesting that regional banks outperform domestic ones.