From d8f0c87ecfff99cb19a419ba7f4f779d6727222c Mon Sep 17 00:00:00 2001 From: hadley Date: Wed, 6 Jul 2016 17:57:03 -0500 Subject: [PATCH] A few more model ideas --- intro.Rmd | 1 + model-basics.Rmd | 2 ++ 2 files changed, 3 insertions(+) diff --git a/intro.Rmd b/intro.Rmd index eb9c999..317be9a 100644 --- a/intro.Rmd +++ b/intro.Rmd @@ -84,6 +84,7 @@ This book focuses exclusively on structured data sets: collections of values tha Exploratory vs. confirmatory +Most people think of models as confirmatory and visualisations as exploratory. But you can have confirmatory visualisations and exploratory models. This book focuses on exploration. ### Formal Statistics and Machine Learning diff --git a/model-basics.Rmd b/model-basics.Rmd index 7607f03..88f57d5 100644 --- a/model-basics.Rmd +++ b/model-basics.Rmd @@ -2,6 +2,8 @@ The goal of a fitted model is to provide a simple low-dimensional summary of a dataset. Ideally, the fitted model will capture true "signals" (i.e. patterns generated by the phenomenon of interest), and ignore "noise" (i.e. random variation that you're not interested in). +A model is a tool for making predictions. Goal of a model is to be simple and useful. + This is a hard problem because any fitted dataset is just the "best" (closest) model from a family of models. Just because it's the best model doesn't make it good. And it certainly doesn't imply that the model is true. But a model doesn't need to be true to be useful. You've probably heard George Box's famous aphorism: > All models are worng, but some are useful.