Degree of Local Polynomials
The local polynomials fit to each subset of the data are almost always of first or second degree; that is, either locally linear (in the straight line sense) or locally quadratic. Using a zero degree polynomial turns LOESS into a weighted moving average. Such a simple local model might work well for some situations, but may not always approximate the underlying function well enough. Higher-degree polynomials would work in theory, but yield models that are not really in the spirit of LOESS. LOESS is based on the ideas that any function can be well approximated in a small neighborhood by a low-order polynomial and that simple models can be fit to data easily. High-degree polynomials would tend to overfit the data in each subset and are numerically unstable, making accurate computations difficult.
Read more about this topic: Local Regression
Famous quotes containing the words degree of, degree and/or local:
“Essential truth, the truth of the intellectualists, the truth with no one thinking it, is like the coat that fits tho no one has ever tried it on, like the music that no ear has listened to. It is less real, not more real, than the verified article; and to attribute a superior degree of glory to it seems little more than a piece of perverse abstraction-worship.”
—William James (1842–1910)
“Nothing in life possesses value except the degree of power—assuming that life itself is the will to power.”
—Friedrich Nietzsche (1844–1900)
“This is the only “wet” community in a wide area, and is the rendezvous of cow hands seeking to break the monotony of chuck wagon food and range life. Friday night is the “big time” for local cowboys, and consequently the calaboose is called the “Friday night jail.””
—Administration in the State of Texa, U.S. public relief program (1935-1943)