Fast model‐based ordination with copulas

Visualising data is a key step in analysis, allowing researchers to find patterns, and assess communicate the results of statistical modelling. In ecology, visualisation often challenging when there are many variables (often for different species or other taxonomic groups) they not normally distributed counts presence–absence data). Ordination common powerful way overcome this hurdle by reducing from response just two three, be easily plotted. traditionally done using dissimilarity-based methods, most commonly non-metric multidimensional scaling (nMDS). last decade, however, model-based methods unconstrained ordination have gained popularity. These primarily based on latent variable models, with estimating underlying, unobserved ecological gradients. Despite some major benefits, drawback their speed, as typically take much longer return result than especially large sample sizes. We introduce copula ordination, new, scalable approach ordination. This method has all desirable properties added advantage that it computationally far more efficient. particular, simulations show an order magnitude faster current can even nMDS sizes, while being able produce similar plots trends these methods.