At EnviroStats Solutions, it is our goal to apply the most appropriate and up-to-date analyses and visualizations to environmental data. We are highly proficient in traditional statistical approaches as well as Bayesian statistics. Our day-to-day modeling consists of:

  • Linear models
  • Generalized linear models (Binomial, Poisson, Gamma, etc.)
  • Beta regression models
  • Linear mixed effects models
  • Generalized linear mixed effects models
  • Zero and one inflated models
  • Classification and regression tree analyses
  • Multivariate analyses, incl. principal component analysis, linear discriminant analysis, non-linear multidimensional scaling techniques
  • Machine learning algorithms

However, before any statistical analyses can take place, all data needs to be prepared and manipulated first. This includes:

  • data checking and cleaning
  • data transformations
  • creating new variables
  • ensuring that the data can be read by the given statistical software package

This step not only represents the most time consuming aspect of any analysis, it is also the entry point for programming errors and typos and could, in the worst case, lead to false conclusions. At EnviroStats Solutions we are aware of this and with our passion to develop clean and concise analyses for our customers, we can help prevent the introduction of errors and typos as well as increase the efficiency of the data analysis workflow.