Project title: On normal approximations to discrete distributions.

Adviser: Steven From

Description: The normal distribution is a continuous distribution often used to approximate

other distributions, including the binomial, Poisson and other discrete distributions. We shall

consider various continuity corrections besides the traditional ‘one-half’ correction discussed

in most textbooks on probability/statistics. Some transformations to normality will also be

considered.

The student will be expected to have had a mathematical (calculus-based) course in probability

and statistics such as Math4740/8746 or Stat 3800/8805. Some knowledge of elementary real analysis is also helpful.