A perceptron for detecting the preferential sampling of locations and times chosen to monitor a spatio-temporal process

The preferential sampling of locations chosen to observe a spatio-temporal process has been identified as a major problem across multiple fields. Predictions of the process can be severely biased when standard statistical methodologies are applied to preferentially sampled data without adjustment. Detecting preferential sampling is currently a technically demanding task. As a result, the problem is often ignored in data analyses. This paper offers a general, intuitive, and computationally-fast solution. A novel approach for testing if a spatio-temporal dataset was preferentially sampled is presented. We refer to the test as a perceptron as it attempts to capture the numerous factors behind the human decision-making that selected the sampled locations and times. Importantly, the method can also help with the discovery of a set of informative covariates that can sufficiently control for the preferential sampling. The discovery of these covariates can justify the continued use of standard methodologies. A thorough simulation study is presented to demonstrate both the power and validity of the test in various data settings. The test is shown to attain high power for non-Gaussian data with sample sizes as low as 50. Finally, two previously-published case studies are revisited and new insights into the nature of the informative sampling are gained. The test can be implemented with the R package PStestR. (C) 2021 Elsevier B.V. All rights reserved.