Generating and Evaluating Predictions with PLS Path Modeling

PLS 路徑模型之產生與預測評估

滸安(Juan Manuel Velasquez Estrada)

指導教授 ： 雷松亞 ； 徐茉莉

國立清華大學/科技管理學院/國際專業管理碩士班/碩士(2015年)

摘要

Partial Least of Squares Path Modeling (PLS-PM) has become a highly utilized statistical tool for business research in recent years. Its flexibility, with no distribution assumptions and its capacity of working with small sample size are often cited as the major characteristics that draw the attention of researchers. Its predictive nature is often cited as one of its more distinctive characteristics, despite the fact that most researchers utilize it only for explanatory purposes. The lack of a formalized algorithm for prediction using PLS-PM models has contributed to the slow development of the technique as a predictive method. In this dissertation we present a suggested algorithm to generate predictions using PLS-PM models, we provide a software implementation as well as a benchmark comparison of its predictive validity against one of the most traditional predictive tools, linear regression. It is then the aim of this dissertation to encourage further research on the subject of PLS-PM as a predictive tool combined with its already known explanatory capabilities, filling the gap in the explanatory-predictive gamut with a reliable method to perform theory informed predictions.

關鍵字

PLS-PM ； Prediction ； Evaluation ； Algorithm ； Explanatory ； Models

並列摘要

Partial Least of Squares Path Modeling (PLS-PM) has become a highly utilized statistical tool for business research in recent years. Its flexibility, with no distribution assumptions and its capacity of working with small sample size are often cited as the major characteristics that draw the attention of researchers. Its predictive nature is often cited as one of its more distinctive characteristics, despite the fact that most researchers utilize it only for explanatory purposes. The lack of a formalized algorithm for prediction using PLS-PM models has contributed to the slow development of the technique as a predictive method. In this dissertation we present a suggested algorithm to generate predictions using PLS-PM models, we provide a software implementation as well as a benchmark comparison of its predictive validity against one of the most traditional predictive tools, linear regression. It is then the aim of this dissertation to encourage further research on the subject of PLS-PM as a predictive tool combined with its already known explanatory capabilities, filling the gap in the explanatory-predictive gamut with a reliable method to perform theory informed predictions.

並列關鍵字

PLS-PM ； Prediction ； Evaluation ； Algorithm ； Explanatory ； Models

參考文獻

Churchill Jr., G.A. (1979). "A Paradigm for Developing Better Measures for Marketing Constructs". Journal of Marketing Research, 16:1, pp. 64-73.

Dijkstra, T. (1983). “Some comments on maximum-likelihood and partial least squares methods.” Journal of Econometrics, (22), 67-90.

Dijkstra, T. (2010). “Latent variables and indices: Herman Wold’s basic design and Partial Least Squares.” In E. Esposito Vinzi et al. (eds.) Handbook of Partial Least Squares. Springer-Verlag: Berlin, pp. 23-46.

Gefen, D., Rigdon, E. E., and Straub, D. W. 2011. “Editor’s Comment: An Update and Extension to SEM Guidelines for Administrative and Social Science Research,” MIS Quarterly (35:2), pp. iii-xiv.

Hair, J. F., Hult, J. G. T. M., Ringle, C. M., & Sarstedt, M. (2014). A Primer on Partial Least Squares Structural Equation Modeling (PLS-SEM).