A sample, in statistics, is a set of observations taken from a larger group, the population, in order to learn about that group without measuring every member. Quantities calculated from a sample, such as the sample mean, are called statistics and serve as estimates of the corresponding population values, which are called parameters. The number of observations is the sample size, written . Larger samples give more precise estimates, provided the sample was selected fairly, ideally by random sampling.
In finance, any set of historical prices or returns is a sample. The population is the full range of outcomes the market could have produced, of which history is a single draw. Ten years of daily returns therefore support only limited conclusions about a strategy's true average return. Financial samples are also rarely random: a database that omits delisted companies suffers from survivorship bias, and a period chosen because it looks favourable misrepresents the whole.