Cross validation

- [Instructor] In the last video we saw…a few classical analytics have it techniques…to have it with goodness of fit,…and to compare models based on their…explanatory power and simplicity.…Now I want to show you how to implement…a much simpler strategy known as Cross Validation.…Which is used in machine learning to compare models.…We divide the data into a training set…which we use to feed the model,…and a testing set which we use to…evaluate the models prediction error.…

So instead of concentrating on in-sample error…as we do with classical techniques,…we will look at how to sample prediction error.…Models and accounts look better by…over fitting the data they're trained on.…Instead they need to in some sense…understand something about the world.…I have all ready included code to load our data set.…I import packages.…And since we will be splitting our data…I have refactored the plotting…so it works on arbitrary data.…

I have also copied the model formulas alone…from the last two videos.…To divide up our data we first shuffle it…