Violation of Independence – Part II
In the previous chapter, we discussed violation of independence for measurements taken repeatedly over time and how temporal correlation structures can be added to linear regression and additive models. We used a regular spaced data set. In this chapter, we consider data measured at multiple spatial locations, and we show how similar correlation structures can be used. The ‘Part II’ in the title refers to irregular spaced data, either in space, time, or along an age or depth gradient. The general principle with spatial data is that things that are close to each other are likely to be more similar than things that are further apart (Tobler, 1979).