Problem Setup
Consider the problem of estimating a deterministic (not Bayesian) parameter from noisy or corrupt data related through the conditional probability distribution . Our goal is to find a "good" estimator for estimating the parameter, which minimizes some given risk function . Here the risk function is the expectation of some loss function with respect to . A popular example for a loss function is the squared error loss, and the risk function for this loss is the mean squared error (MSE).
Unfortunately in general the risk cannot be minimized, since it depends on the unknown parameter itself (If we knew what was the actual value of, we wouldn't need to estimate it). Therefore additional criteria for finding an optimal estimator in some sense are required. One such criterion is the minimax criteria.
Read more about this topic: Minimax Estimator
Famous quotes containing the word problem:
“Give a scientist a problem and he will probably provide a solution; historians and sociologists, by contrast, can offer only opinions. Ask a dozen chemists the composition of an organic compound such as methane, and within a short time all twelve will have come up with the same solution of CH4. Ask, however, a dozen economists or sociologists to provide policies to reduce unemployment or the level of crime and twelve widely differing opinions are likely to be offered.”
—Derek Gjertsen, British scientist, author. Science and Philosophy: Past and Present, ch. 3, Penguin (1989)