Garden with Insight
Product area
Help System
Contents
Quick start
Tutorial
How-to
Models
|
Garden with Insight v1.0 Help: empirical/theoretical
An empirical model is one based primarily on observation. An
empirical model uses functions that best fit existing data to predict future values -- it is data-
centric. For example, if you notice that your tomato seedlings are wilting and want to make an
empirical model about it, you might remember that last year this happened and that watering the plants
brought them back. You don't necessarily care why watering the plants restored their health: you
only care that it worked.
A theoretical model is one based primarily on what are called first principles, or scientific laws
that have been created by theorizing and testing ideas about how some process works. An empirical model
uses functions that best fit existing explanations to predict future values -- it is idea-centric. For
example, if you want to make a theoretical model about why your tomato seedlings are wilting, you would
consider all of the reasons tomatoes wilt and decide that the most likely problem (from the relative
importance of all the causes) is that they are lacking in adequate water.
Of course models can have both empirical and theoretical aspects, and there is a broad range between
these two extremes. EPIC (the model on which this simulation is based) is mainly empirical. The parameters for the many S curves that
drive the simulation are based on published experiments, not on theories (though they may mesh with
theoretical explanations). EPIC bases its predictions on results, not on reasons. Most models that cover
this broad a range are empirical, because the processes involved are very complex. Theoretical models
tend to examine a much smaller phenomenon, such as only photosynthesis or only ground water flow.
|