Suppose I have a model like brm(y ~ s(x1) + s(x2) + x3) . The decision makers who will be using the model would like to see how predictions come together for certain observations. To help facilitate this, I would like to see the contribution of each of the additive components to the overall predicted value. In other words, I would like to be able to decompose a prediction into the contribution from each of the terms in the model. Is this possible to do with brms?
What you could do is compare the whole model to the model leaving out a specific term and then investigating how much the predictions change / get worse.