Cupar Analytics Ltd thrives on working with complex, large volume datasets, that are often disorganised and repleat with human-error, providing context, order and new meaning. This is achieved through statistical methods, machine learning and AI, and the use (and develoment) of software and applications.
Statistics (Bayesian and frequentist methods)
Distance sampling, occupancy modelling, N-mixture models, spatial capture-recapture methods, population viability analysis, regression models (e.g. GLMs, GAMs, GLMMs), state-space models, numerical and analytical modelling.
Machine learning and AI
Custer analysis, neural networks, random forests, and image processing.
Software and apps
Rshiny apps, Django python apps, Dockerisation of apps, publication of web services, Microsoft Azure apps, Databricks, Echoview, GenEst, EoA software, and Vortex.