id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
work_pxefz433avb75cjcc3rayffpc4	Guillaume Rochefort-Maranda	Simplicity and model selection	2016	32	.pdf	application/pdf	6347	683	69	a data set and explain how we can choose a regression model with an In order to construct a data set, I first define the following function f (x): However, both regressions are different in the sense that a polynomial regression is a parametric model and a kernel regression is a nonparametric Finally, we need to figure out the number of parameters p that will determine the best estimate for f (x) out of all the possible polynomial regression models that can fit the data set. our data set; construct a model; and then compute the square of the difference between our prediction of the removed observation and that observation. values of its adjustable parameters that best fit the data set. As mentioned in section 2.1, both the polynomial and the kernel regression estimates by on the observed data (x, y) can be defined with a linear	./cache/work_pxefz433avb75cjcc3rayffpc4.pdf	./txt/work_pxefz433avb75cjcc3rayffpc4.txt