Abstract

This chapter reviews some of the problems typical of data analyses in industry. In particular, this chapter will investigate problems associated with fitting a probability density function (pdf) to data when only a small sample of data is available (Alderman and Mense (2007)). This is never a matter of choice, but is often necessitated by engineering practice, e.g., when a design may still be changing or at any time that data collection may be limited by availability or cost. Having hundreds of data points is best, but in practice this may not be possible.