piece-wise linear functions, also known as broken-stick models, can be used to account for nonlinear relationships with more easily interpretable regression parameters. In our use of linear splines, we aim to capture two different types of nonlinear relationships: unimodal, where an increasing trend is followed by a decreasing one, or vice versa; or ‘saturation curve’, by which we mean a monotonic relationship that appears to flatten for increasing values of the covariate.
← all excerpts
Model building and assessment of the impact of covariates for disease prevalence mapping in low-resource settings: to explain and to predict.
1
—
—