Till startsida
To content Read more about how we use cookies on gu.se

Making the predictive model

THIS IS WHERE OUR USERS SHOULD FIND THE MOST IMPORTANT THINGS TO CONSIDER WHEN ORDERING OR EVALUATING A MAP FROM A CONSULTANT

The choice of modelling method should be based on:

  • composition of biological data - reliable presences and absences or presences only?
  • relationship between response variable and environmental/predictor data - linear or nonlinear?
  • interactions between environmental variables/predictor data – absent or present?

Four appropriate methods
Results from PREHAB indicate that the following methods are useful for determining the statistical relationship between response variable and predictor data, i.e. finding/making your model for predicting vegetation distribution:

  • Generalized additive models (GAM),
  • Multivariate adaptive regression splines (MARS),
  • Random forest (RF) and
  • Maximum entropy modelling (MaxEnt)

Generally, the obtained results from PREHAB indicate that good to excellent classification accuracy of vegetation distribution by these methods is highly probable, where data consists of more than 400 samples and 30% presences.

 


Cover issues, such as:
• Advice: (important points, "warnings")
• General discussion of the methods: present the methods, basics and their pros and cons
• PREHAB results (as examples)

Choice of modelling method

When selecting an appropriate stochastic modelling method there are a number of factors that should be considered:

  • which type of biological data that is available?
  • info of geographical? coordinates,
  • availability of secondary information such as environmental variables?

More

PREHAB evaluation

PREHAB selected four methods for statistical modelling of vegetation distribution – GAM, MARS, RF and MaxEnt – and tested their performance in four case-study areas.
More

© University of Gothenburg, Sweden Box 100, S-405 30 Gothenburg
Phone +46 31-786 0000, About the website

| Map

The University of Gothenburg uses cookies to provide you with the best possible user experience. By continuing on this website, you approve of our use of cookies.  What are cookies?