Resemblance profiles as clustering decision criteria: Estimating statistical power, error, and correspondence for a hypothesis test for multivariate structure

Abstract

Clustering data continues to be a highly active area of data analysis, and resemblance profiles are being incorporated into ecological methodologies as a hypothesis testing‐based approach to clustering multivariate data. However, these new clustering techniques have not been rigorously tested to determine the performance variability based on the algorithm's assumptions or any underlying data structures. Here, we use simulation studies to estimate the statistical error rates for the hypothesis test for multivariate structure based on dissimilarity profiles (DISPROF). We concurrently tested a widely used algorithm that employs the unweighted pair group method with arithmetic mean (UPGMA) to estimate the proficiency of clustering with DISPROF as a decision criterion. We simulated unstructured multivariate data from different probability distributions with increasing numbers of objects and descriptors, and grouped data with increasing overlap, overdispersion for ecological data, and correlation among descriptors within groups. Using simulated data, we measured the resolution and correspondence of clustering solutions achieved by DISPROF with UPGMA against the reference grouping partitions used to simulate the structured test datasets. Our results highlight the dynamic interactions between dataset dimensionality, group overlap, and the properties of the descriptors within a group (i.e., overdispersion or correlation structure) that are relevant to resemblance profiles as a clustering criterion for multivariate data. These methods are particularly useful for multivariate ecological datasets that benefit from distance‐based statistical analyses. We propose guidelines for using DISPROF as a clustering decision tool that will help future users avoid potential pitfalls during the application of methods and the interpretation of results.

Strategic categories and competition: significant clustering for strategic groups

“…Given the problems associated with the dreaded duality, it might seem surprising that these tests are quickly gaining popularity in the field of ecology (Kilborn et al , 2017). It turns out that these issues are not terribly relevant in that field.…”

Fluctuating fishing intensities and climate dynamics reorganize the Gulf of Mexico's fisheries resources

“…, Kilborn et al. ). This form of clustering, referred to as SIMPROF clustering hereafter, assesses H o2 = “there is no multivariate structure among objects (years) with respect to the set of descriptors (Y).” The method was developed as a form of hypothesis testing‐based clustering (Clarke et al.…”

“…) and is well suited for high‐dimensional, continuous datasets with relatively large sample sizes (Kilborn et al. ). The assessment of H o2 was made at all possible levels of resemblance identified by an UPGMA clustering solution produced from a Euclidean resemblance matrix (Y Euc ; Legendre and Legendre ).…”

Retention and export of planktonic fish eggs in the northeastern Gulf of Mexico

“…Egg abundance was square root transformed, and Bray-Curtis similarity was calculated for all possible station pairs. Cluster analysis was performed using these Bray-Curtis similarities, and a similarity profile analysis (SIMPROF) was used to identify statistically significant groupings of stations within the results of the cluster analysis (Kilborn et al, 2017). A seriated heatmap was generated to allow simultaneous visual comparisons of (1) station compositional similarity and (2) species associations.…”