Question, is it easier to set records earlier in a record than later in a record? To put it another way, would you expect a record to have more weight in a small sample, or a large sample? :lol:
There is a reason that it is better to look at the distribution of events, and to normalize by the local standard deviation, as James Hansen and colleagues did. For one it minimizes the bias that comes from comparing records in a small sample to a large one.
There is a reason that it is better to look at the distribution of events, and to normalize by the local standard deviation, as James Hansen and colleagues did. For one it minimizes the bias that comes from comparing records in a small sample to a large one.