A population, in statistics, is the complete set of items, individuals or events about which a conclusion is sought. It is contrasted with a sample, the subset that is actually observed. Numbers that describe a population, like the population mean and the population standard deviation, are called parameters. They are usually unknown and are estimated from a sample, ideally drawn by random sampling.
In some financial questions the population is concrete, such as all the shares in an index or all the loans in a bank's book. More often it is abstract. When an analyst studies a strategy's daily returns, the population is every return the underlying process could produce, including those in the future, and recorded history is one sample from it. The process may change over time, so past and future may belong to different populations, and the sample may be unrepresentative, as with survivorship bias, where failed funds or companies have dropped out of the data.