Table 7

Predictive power assessment

PLS-predict out-of-sampleQ2 predictPLS-SEM_ RMSEPLS-SEM_MAELM_RMSELM_MAEError Dif. RMSE PLS-LMError diff. MAE PLS-LMAbsolute values of skewness
Assumption0.1460.9310.7620.9120.7340.0190.028−0.654
Proactivity0.2030.9030.7450.9500.780−0.047−0.035−0.407
Trend0.3170.8360.6870.8620.699−0.026−0.012−0.312
Autonomy0.2010.9030.6960.9010.7090.002−0.013−0.220
Recognition0.3810.7930.6380.8290.658−0.036−0.020−0.576
Support0.3740.7990.6400.7960.6360.0030.004−0.585
Time0.3130.8390.6740.8640.703−0.025−0.029−0.292
ECS0.2760.8580.6830.8820.698−0.024−0.015−0.484
ENS0.1220.9440.7490.9660.756−0.022−0.007−0.916
SOCS0.2600.8670.7050.8640.7070.003−0.002−0.469
 1st. Step 2rd. Step 3nd. Step

Note(s): PLS_predict: K = 4; n = 30; 1st. Step: Q2 predict >0; 2nd. Step: error difference RMSE – PLS LM < 1 for most indicators; 3rd. Step: absolute value of skewness <1. Abbreviations: ECS: economic sustainability; ENS: environmental sustainability; SOCS: social sustainability; MAE: mean absolute error and PLS-LM: difference between the PLS-SEM model and the linear model (LM)

Source(s): Authors' own work

or Create an Account

Close subscription notice
Close access options