Forecasting Corporate Failure in the Chinese Energy Sector: A Novel Integrated Model of Deep Learning and Support Vector Machine

Abstract

Accurate forecasts of corporate failure in the Chinese energy sector are drivers for both operational excellence in the national energy systems and sustainable investment of the energy sector. This paper proposes a novel integrated model (NIM) for corporate failure forecasting in the Chinese energy sector by considering textual data and numerical data simultaneously. Given the feature of textual data and numerical data, convolutional neural network oriented deep learning (CNN-DL) and support vector machine (SVM) are employed as the base classifiers to forecast using textual data and numerical data, respectively. Subsequently, soft set (SS) theory is applied to integrate outputs of CNN-DL and SVM. Hence, NIM inherits advantages and avoids disadvantages of CNN-DL, SVM, and SS. It is able to improve the forecasting performance by taking full use of textual data and numerical data. For verification, NIM is applied to the real data of Chinese listed energy firms. Empirical results indicate that, compared with benchmarks, NIM demonstrates superior performance of corporate failure forecasting in the Chinese energy sector.

Assessment of a failure prediction model in the European energy sector: A multicriteria discrimination approach with a PROMETHEE based classification

“…This section deals with one of the initial steps in the development of a failure prediction model: the selection of the most predictive variables. It consists of the review of the distress prediction literature with special attention to the most predictive variables employed in the energy sector (see for a literature review of FRs on failure prediction models: Xu et al, 2019;Liang et al, 2016;Du Jardin, 2016). Table 2 shows the set of 42 FRs derived from the literature review, which are classified according to six firms' dimensions, together with their acronyms and definitions.…”

Evaluating the Sustainability of Microfinance Institutions Considering Macro-Environmental Factors: A Cross-Country Study

“…In the traditional TOPSIS, the weight of each criterion is determined by the expert system method [54]. The expert system method relies heavily on expert knowledge and the ability to be widely employed [44,55]. To overcome the disadvantages of traditional TOPSIS, we implement a new TOPSIS approach by integrating soft set theory and traditional TOPSIS.…”

Approaches to visualizing big text data at the stage of collection and pre-processing

“…The work related to prediction incidents related to the Chinese energy market is also interesting. [20] The addition of text data in a predictive model also gave the increase prediction accuracy. There are also works that indicate that for a number of tasks adding textual data to the analysis will not increase the accuracy of prediction, For example, the text data added to the model for prediction success in obtaining investments for companies did not increase the accuracy of the model.…”