Partial least squares path modeling

The Partial Least Squares (PLS) path modeling approach to structural equation modeling (SEM), is a component-based estimation procedure different from the covariance-based Structural equation modeling approach. Unlike the covariance-based approach to structural equation modeling, PLS-PM does not aim at reproducing the sample covariance matrix. It is more oriented towards optimizing explained variance (prediction) rather than statistical accuracy of the estimates

The PLS structural equation model is composed of two sub-models: the measurement model and structural model. The measurement model composed of the relationships between the observed data and the latent variables. The structural model is represents the relationships between the latent variables.

In PLS path modeling an iterative algorithm solves the structural equation model by estimating the latent variables by using the measurement and structural model in alternating steps, hence the procedure's name, partial. The measurement model estimates the latent variables as a weighted sum of its manifest variables. The structural model estimates the latent variables by means of linear regression between the latent variables estimated by the measurement model. This algorithm repeats itself until convergence is achieved.