Hybrid Corporate Performance Prediction Model Considering Technical Capability

Abstract

Abstract:Many studies have tried to predict corporate performance and stock prices to enhance investment profitability using qualitative approaches such as the Delphi method. However, developments in data processing technology and machine-learning algorithms have resulted in efforts to develop quantitative prediction models in various managerial subject areas. We propose a quantitative corporate performance prediction model that applies the support vector regression (SVR) algorithm to solve the problem of the overfitting of training data and can be applied to regression problems. The proposed model optimizes the SVR training parameters based on the training data, using the genetic algorithm to achieve sustainable predictability in changeable markets and managerial environments. Technology-intensive companies represent an increasing share of the total economy. The performance and stock prices of these companies are affected by their financial standing and their technological capabilities. Therefore, we apply both financial indicators and technical indicators to establish the proposed prediction model. Here, we use time series data, including financial, patent, and corporate performance information of 44 electronic and IT companies. Then, we predict the performance of these companies as an empirical verification of the prediction performance of the proposed model.

Deep Learning-Based Corporate Performance Prediction Model Considering Technical Capability

“…The RMSE values between the actual and predicted corporate performance values for the models, using the same test data, are shown in Table 3. In general, the prediction performance of SVM-based models is superior to that of neural network-based models in forecasting data with high volatility such as net profit and operating profit [3,7,32,33,35,37]. However, as shown in Table 3, the prediction performance of the proposed DBN-based model is the most accurate.…”

“…Then, the network parameters are fine-tuned by applying a backpropagation algorithm, using relatively recent training data. Using recent data in the fine-tuning process means recent trends are given priority in predictions, thus improving the accuracy of predictions of time-series data (e.g., corporate performance) [37,42]. In addition, the model is expected to have sustainable prediction performance if new data are periodically added to the training data used in the fine-tuning phase.…”

“…The descriptive statistics of the four patent indicators are shown in Table 2. In many cases, it takes a number of years to commercialize a technology after applying for a patent and, thus, improve corporate performance [37]. Therefore, we conduct a correlation analysis to determine the time-lag that maximizes the correlation coefficient between technical indicators and corporate performance.…”

“…Hsu et al (2009) predict the stock price of a firm using a SVR and a self-organizing map [3]. Lee et al (2016) proposed a SVR-based corporate prediction model that searches for optimal SVR parameters using a genetic algorithm [37]. Hinton et al (2006) proposed a DBN that can train deep neural networks effectively using a RBM and the wake-sleep algorithm [11].…”

“…The descriptive statistics of the 11 financial indicators are shown in Table 1. In recent years, there have been studies using technical indicators that represent the technological capability of a company as predictors to predict corporate performance [16][17][18]37]. Many of these studies apply patent data to derive the technical indicators [16][17][18][19][20][21][22]37].…”