(i) Example from Histogramming (Physics) Events---III

• Also,
• And so error of estimate is
• where we can estimate
• the number of events in histogram.
• Actually, we can ``derive'' this result more simply using just the single indicator variable on the events that fall in the histogram.
• This is a little flaky because the value is a random variable (with an error ) as it comes from selecting events from the larger sample N.
• The tortuous argument above can convince you that it doesn't matter that is a random variable!
• The naive argument which gives correct answer is

  

• and so the error is in agreement with above ( is standard deviation of ).