The traditional measurement of active management is to calculate the tracking error of the fund; as measured by the standard deviation of the difference in a fund's returns versus its benchmark returns. However, tracking error alone is an inadequate measure of fund activity since even very actively managed funds can in fact generate rather low tracking errors. Therefore judging the activity level of a fund based solely on tracking error can be misleading. In addition, while tracking error volatility is easy to calculate, it only infers what the manager is doing at the portfolio level and does tell you how the tracking errors were generated. For instance, Cremers and Petäjistö (2009) argue that the two distinct approaches to active management, stock selection or factor timing, can produce significantly different tracking errors. Therefore, instead of using tracking error alone, Cremers and Petäjistö (2009) suggest that a more comprehensive picture of active management can be achieved by including Active Share into the calculations, where the Active Share of the fund is measured by the weighted difference between the fund’s holdings and those of the benchmark portfolio. In other words, Cremers and Petäjistö (2009) claim that using Active Share and tracking error together enables investors to distinguish between the types of active management used by funds, stock selection and factor timing, and so identify whether the fund is using one of the following active management approaches: diversified, concentrated, factor bets, moderately active and closet indexers.