View on GitHub

MT_ASA

2022 MT ASA Chapter Meeting

Please register for the meeting. Registration is free and we will be providing lunch for participants (at Rendezvous Dining Hall) but please register for the event.

Event Schedule

Unless otherwise noted, events will occur in room 166 of American Indian Hall on the campus of Montana State University.

Friday October 28

Panelists

Speakers

Megan Dailey Higgs, Critical Inference.

Title: Pausing to take a deeper look at assumptions

Abstract: Assumptions are a necessary part of statistical modeling and making inferences. Formal training in Statistics tends to encourage a very mathematical treatment of assumptions, with relatively low expectations for justifying the reasonableness of assumptions or adequately conveying the extent to which conclusions are conditional on assumptions. Given that statistical inferences are often used to support scientific inferences and decision-making, it is important to acknowledge the conditional aspect of results, as well as take time to pause and more deeply consider the layers of assumptions that make up the foundation of an analysis. I believe statisticians have a responsibility to better convey the conditional nature of results, and to work with subject matter experts to translate and interrogate assumptions within a particular scientific context. This is a serious scientific challenge with huge implications in terms of understanding and conveying appropriate use of, and trust in, statistical results. The goal of this high-level talk is to increase (or re-highlight) awareness and spur discussion about related challenges and strategies.

Ian Laga, Montana State University

Title: A case-control sampling strategy for zero-inflated models with an application to female sex worker mapping in sub-Saharan Africa

Abstract: In this work, we propose a subsampling procedure to decrease computation time of zero-inflated models. Zero-inflated models are increasing in popularity and are vital to a variety of applications and disciplines. Performing variable selection, estimating parameters, and diagnosing model fit for zero-inflated models is often prohibitively slow, especially for large data sets and Bayesian models. We show that we can consistently estimate the intercept and slope parameters of both the zero and conditional models. Performance is evaluated using a spatial presence-only data set related to the number of female sex workers in four countries in sub-Saharan Africa.

John Smith, Montana State University

Title: Predicting Carbon in Forest Ecosystems: A Case for Process-based Modeling Approaches

Abstract: Approaches to prediction and forecasting in ecological applications broadly fall into three categories: time-series approaches, space-for-time approaches, and process-based modeling approaches. Empirical statistical models built using time-series and space-for-time approaches may be limited in what type of dynamics they are able to capture: i.e. time-series approaches may accurately capture the dynamics of short term processes and drivers, but can miss out on capturing dynamics of long term processes that are occurring on the order of decades to centuries. Process-based models, on the other hand, are mathematical representations of the processes that govern a system. These models combine our knowledge of the biological processes of the system with statistical tools for uncertainty quantification. In theory, process-based models are able to capture dynamics across all time-scales, and will lead to better forecasts and predictions over any horizon. In practice there are many difficulties, including overparameterization and identifiability, enforcing constraints of the underlying system in a statistically coherent manner, and finding suitable fitting methods. We examine these difficulties and provide possible solutions through the lens of a case study: predicting Leaf Area Index at the University of Notre Dame Environmental Research Center.

Kevin Surya, Montana State University

__Title:__The Genomic Evolution of Living Fossils

Abstract: Most species appear radically different from their ancestors that lived hundreds of millions of years ago. Yet a few species are remarkably similar to their ancient ancestors. Charles Darwin called these ancient-looking creatures living fossils. Perhaps the most mesmerizing of which is the coelacanth. This species dwells off the eastern coast of Africa and Indonesia and appears morphologically similar to 300 million-year-old fossils. The genomes of several living fossils have been sequenced recently, revealing low average evolutionary rates in their protein-coding DNA sequences. Genomic and anatomical bradytely (extremely slow evolution) are poorly understood, especially in the era of big data. For example, how can animal genomes evolve slowly despite continual environmental change? We are currently unraveling this question with Bayesian evolutionary models. Elucidating what drives the pace of living fossils’ genomic evolution is a central problem in evolutionary theory and is essential for protecting species in today’s world of unprecedented climate change.