The x-calculated values are linear combinations of the x-scores. The x-calculated values contain the variance in the terms explained by the PLS regression model. Observations with relatively small x-calculated values are outliers in the x-space and are not well explained by the model.
The x-calculated matrix, similar to the original x-matrix, is an n x p matrix, where n = number of observations and p = number of terms. The x-calculated values are on the same scale as the predictors.
If the number of components equals the number of terms, then x-calculated value equals the original x-value.