For example, a PCA run on bond market data using Treasury bonds across various maturities usually shows that the first (most important) risk factor statistically corresponds to the level of interest rates, or what a theory‐driven risk model might call interest rate risk. PCA and other statistical models are commonly used in equity markets as well, and these models typically find that the market itself is the first, most important driver of returns for a given stock, usually followed by its sector. These statistical risk models are most commonly found among statistical arbitrage traders, who are betting on exactly that component of an individual stock’s returns that is not explained by systematic risks. It is important to note that such statistical methods may discover entirely new systematic risk factors, which a reasonable observer might be inclined to acknowledge exist but for which names have not been assigned. On the other hand, statistical risk models are subject to being fooled by the data into finding a risk factor that will not persist for any useful amount of time into the future. It is also possible for a statistical risk model to find spurious exposures, which are just coincidences and not indicative of any real risk in the marketplace. This is a delicate problem for the researcher.