All three must be incorporated. The absence of each has been used to criticize research on the EMH. Some studies used information that was not available to investors at the time the historical data was evaluated. Many studies did not explicitly account for transaction costs, even though these would clearly reduce profits. Finally, risk adjustments must be made. As an example, consider a trading strategy that looks at percentage price breakouts as described in the text. Large percentage price breakouts are more likely to be found in lower-priced stocks. Many of these are riskier, smaller firms, so a test like this would be biased toward riskier stocks and should require a higher return just to match the market.