A wage is one of key factors enabling economic expansion. It is related to many economic agents such as employees, entrepreneurs and the rest of the economy. Therefore, the behavior of entrepreneur’s wage determination should be thoroughly considered. This research aims to study factors influencing the determination of wages in Thai labor market during 2007 – 2017 by using nationwide official data of Household Socio-Economic Survey (SES) and Labor Force Survey (LFS). This study also formulates the forecasting model of Thai labor wages by using the regression method and Artificial Intelligence (AI) techniques, which are the Artificial Neural Network (ANN) framework and the Random Forest (RF) algorithm.
2
The outcome of this study indicates that determinants influencing the wage determination in the last 10 years comprise of three main factors, which are (1) personal attributes, (2) career and business characteristics and (3) geographical indictors. Moreover, it is found that Random Forest algorithm is the most accurate forecasting method, and it also allows the computation of degree of significance of each factor. Although ANN framework is more accurate than the regression method, there are limitations in its ability to explain the detail of causality and there is still a limitation in processing dataset that includes many discrete variables. These findings suggest the future extension of applying artificial intelligence techniques in examining and forecasting key economic indicators.
Keywords: Artificial intelligence, Human resource management, Wages Forecasting, Neural Network algorithm, Random Forest algorithm, Thai economy
(2) ในการศึกษาต่อไป อาจจะทำการทดสอบ Robustness Check ของค่าระดับความแม่นยำของแต่ละอัลกอริทึม โดยการปรับสัดส่วน Training Data และ Testing Data จาก 70:30 เป็น 50:50 หรือ 40:60 เพื่อหลีกเลี่ยงปัญหา Overfitting และเพื่อให้แน่ใจอัลกอริทึมที่ถูกเทรนแล้วสามารถนำไปใช้พยากรณ์กับข้อมูล Out of Sample หรือข้อมูลที่เกี่ยวข้องแม่นยำมากขึ้น
Mincer, J. (1974). Schooling, Experience and Earnings. New York: National Bureau of Economic Research.
Bhulai, D. S. (2016). Analysing Which Factors Are of Influence in Predicting. Research Paper Business Analytics Vrije Universiteit Amsterdam.
Breiman L. (2001). Random Forests. Machine Learning, 45(1), 5–32.
Changki, L. & Gary, G. L., (2006). Information Gain and Divergence – base Feature Selection for Machine Learning-based Text Categorization. Journal of Information Processing and Management, 42(1), 155-165.
Gareth J., Daniela W., Trevor H. & Robert T. (2015). An Introduction to Statistical Learning with Applications in R. London: Springer New York Heidelberg Dordrecht.
Gert, B., Shyngys, K. & Jozef, K. (2019). Wage Indexation and Jobs: a Machine Learning Approach. Research Center of Economic: VIVES DISCUSSION PAPER.
Elamparithi, V. (2014). An Efficient Classification Algorithms for Employee. International Journal of Research in Advent Technology, 2(9), 29-32.
Pham, H. T. & Reilly, B. (2007). The Gender Pay Gap in Vietnam, 1993-2002; A Quantile Regression Approach. Journal of Asian Economics 18: 775-808
Quadri, M.& Kalyankar, D.N.V. (2010). Drop Out Feature of Student Data for
Academic Performance Using Decision Tree Techniques. Global Journal of
Computer Science and Technology 10(2), 2-5.
Sebastian, R. (2015). Python Machine Learning. Birmingham,UK: Packt Pulishing.
Tangtipongkul, K., (2015). Rates of Return to Schooling in Thailand. Asian Development Review, 32(2), 38-64,
Kalaivani, M.V. (2014). An Efficient Classification Algorithms for Employee. International Journal of Research in Advent Technology, 2(9), 29-32.
An, Z.Y., Shen, H.F. & Wang, S.T. (2017). Employee Recruitment Method Based on Random Forest . 2017 5th International Conference on Computer, Automation and Power Electronics (CAPE 2017), 182-186.