Swarm Intelligence

This fact suggests that if we take a number of random samples of some

size, compute their means, and plot them, the distribution—called the

sampling distribution of the means—will be narrower where n is larger. Ob-

viously if the samples were as large as the population, they would all

have the same mean, which would equal the population mean, and their

variance would equal zero. Smaller samples will vary around the popula-

tion mean. As you would think, the mean of the sampling distribution of

the means equals the population mean. An interesting finding of this

kind of exercise is that, if the samples contain about 30 or more observa-

tions, the sampling ...