id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
work_g6cnurbjhneavh2kjkzghvejtm	Jan-Willem Romeijn	Inherent Complexity: A Problem for Statistical Model Evaluation	2017	16	.pdf	application/pdf	5236	404	68	This paper investigates a problem for statistical model evaluation, in particular for curve fitting: by employing a different family of curves we can the example, the best fitting curve from the quadratic model will have a higher complex model the same number of data points will be used to determine a data with a model based on trigonometric functions, or sine curves for short. a perfect fit to m data points with a polynomial curve of degree m − 1. Figure 2: The sine curve that perfectly fits the scatter plot. The fact that there are infinitely many equally well-fitting sine curves incapacitates some of the standard model evaluation tools. By contrast, the set of best fitting sine curves alters significantly with the addition of a point, not so much by Well fitting polynomial curves of a given degree are concentrated in a particular region within the model, in which posterior probability	./cache/work_g6cnurbjhneavh2kjkzghvejtm.pdf	./txt/work_g6cnurbjhneavh2kjkzghvejtm.txt