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:
“A certain degree of neurosis is of inestimable value as a drive, especially to a psychologist.”
—Sigmund Freud (1856–1939)
“As to the bride, she is blithe as the month; if one can compare in any degree a weed of December, with the fragrance of May; for a weed in truth it is, and a weed not in its first prime.”
—Frances Burney (1752–1840)
“America is the world’s living myth. There’s no sense of wrong when you kill an American or blame America for some local disaster. This is our function, to be character types, to embody recurring themes that people can use to comfort themselves, justify themselves and so on. We’re here to accommodate. Whatever people need, we provide. A myth is a useful thing.”
—Don Delillo (b. 1926)