A cartoon to teach about the measurement issues of bias, reliability, and validity. Cartoon by John Landers (www.landers.co.uk) based on an idea from Dennis Pearl (The Ohio State University). Free to use in the classroom and on course web sites.
In God we trust, all others bring data. An unsourced quote often attributed to American statistician and quality control pioneer William Edwards Deming (1900-1993). The quote has also been a motto of NASA for several decades.
A cartoon to teach about the distributional assumptions in linear models. Cartoon by John Landers (www.landers.co.uk) based on an idea from Dennis Pearl (The Ohio State University). Free to use in the classroom and on course web sites.
A cartoon to teach about the Margin of error in sample surveys. Cartoon by John Landers (www.landers.co.uk) based on an idea from Dennis Pearl (The Ohio State University). Free to use in the classroom and on course web sites.
The number of degrees of freedom is usually self-evident - except for the analysis of data that have not appeared in a textbook. A quote from M.I.T. professor of management David Durand (1912- 1996) Published in a letter to the editor of "The American Statistician" June, 1970 as part of a tongue-in-cheek "Dictionary for Statismagicians." The quote also appears in "Statistically Speaking: A dictionary of quotations" compiled by Carl Gaither and Alma Cavazos-Gaither.
This applet draws a Gamma process (a stochastic process with independent increments X(s + t) - X(s).) Click the mouse in the window to start zooming. Click again to stop. The total increase occurs at a countable set of jumps. The simulation gives some idea of this.
This applet draws one-dimensional Brownian motion. Click the mouse in the window to start zooming. Click again to stop. Since Brownian motion is self-similar in law, all of the zoomed pictures look the same.
These pages from the University of Melbourne explain statistical concepts using various examples from medicine, science, sports, and finance. The intent is not computational skill but conceptual understanding. Some pages also contain data.
This article introduces Radial Basis Function (RBF) networks. These networks rely heavily on regression analysis techniques. Topics include Nonparametric Regression, Classification and Time Series Prediction, Linear Models, Least Squares, Model Selection Criteria, Ridge Regression, and Forward Selection.
