Prelude # 4

  • Date/time:
    Wednesday, April 27, 2022, 3.00pm to 5.00pm CET (Berlin time zone); via Zoom 
  • Title:
    Predictive model evaluation and comparison in PLS-SEM
  • Instructors:
    Benjamin Liengaard (Aarhus University)

       Nicholas P. Danks (Trinity College Dublin)


Part I

Content from Youtube can't be displayed due to your current cookie settings. To show this content, please click "Consent & Show" to confirm that necessary data will be transferred to Youtube to enable this service. Further information can be found in our Privacy Policy. Changed your mind? You can revoke your consent at any time via your cookie settings.

Consent & Show

Part II

Content from Youtube can't be displayed due to your current cookie settings. To show this content, please click "Consent & Show" to confirm that necessary data will be transferred to Youtube to enable this service. Further information can be found in our Privacy Policy. Changed your mind? You can revoke your consent at any time via your cookie settings.

Consent & Show

CONTENTS 

Methodological research on Partial Least Squares Structural Equation Models (PLS-SEM) is increasingly focused on the ability of the method to evaluate out-of-sample predictive power of models (Legate et al. 2021; Liengaard et al. 2020; Shmueli et al. 2019). Such developments are beneficial for establishing validity, practical utility and generalizability of models. However, further improvement is still needed to strengthen PLS as a causal-predictive method. 

 

This workshop will contain three main parts: 1) a general introduction to prediction and predictive analyses in PLS-SEM (Hair et al., 2021). 2) Guidelines to estimating and evaluating overfit at the construct level and identifying possible sources of overfitting (Danks et al., 2021). 3) A set of out-of-sample prediction tests that compares a given PLS-SEM model to either a naïve benchmark model or linear benchmark model (Sharma et al. 2020).

 

On completion of this workshop, you obtain a general understanding of predictive evaluations in PLS-SEM and will be on par with some of the newest techniques within this area. 

TIME & PLACE 

Wednesday, April 27, 2022, 3.00pm to 5.00pm CET (Berlin time zone); via Zoom

REGISTRATION

Please register before April 20, 2022.


Nicholas P. Danks

Trinity Business School, Trinity College Dublin

https://scholar.google.com/citations?hl=en&user=D7nnR8gAAAAJ

 

Benjamin Liengaard

Department of Economics and Business Economics, Aarhus University

https://pure.au.dk/portal/en/persons/benjamin-liengaard(20b26e34-52de-46c0-bfd7-1fce386755bf).html

REFERENCES

 

  • Danks, N.P., Ray, S., Shmueli, G. (2021). The Composite Overfit Analysis Framework: Assessing the Out-of-sample Generalizability of Construct-based Models Using Predictive Deviance, Deviance Trees, and Unstable Paths. Working Paper.
  • Hair, J.F., Hult, G.T.M., Ringle, C.M., Sarstedt, M., Danks, N.P., Ray, S. (2021). Partial Least Squares Structural Equation Modeling (PLS-SEM) Using R: A Workbook
  • Springer Nature (open access). https://link.springer.com/book/10.1007/978-3-030-80519-7
  • Legate, A. E., Hair Jr, J. F., Chretien, J. L., & Risher, J. J. (2021). PLS‐SEM: Prediction‐oriented solutions for HRD researchers. Human Resource Development Quarterly.
  • Liengaard, B., Sharma, P. N., Hult, G. T. M., Jensen, M. B., Sarstedt, M., Hair, J. F., & Ringle, C. M. (2020). Prediction: Coveted, Yet Forsaken? Introducing a Cross-validated Predictive Ability Test in Partial Least Squares Path Modeling. Decision Sciences, Advance online publication.
  • Sharma, P. N., Liengaard, B. D., Hair, J.F., Sarstedt, M., & Ringle, C. M. (2021). Predictive Model Assessment and Selection in Composite-based Modeling Using PLS-SEM: Extensions and Guidelines for Using CVPAT. Working Paper
  • Shmueli, G., Sarstedt, M., Hair, J. F., Cheah, J., Ting, H., Vaithilingam, S., & Ringle, C. M. (2019). Predictive Model Assessment in PLS-SEM: Guidelines for Using PLSpredict. European Journal of Marketing, 53(11), 2322-2347.