Abstract

This chapter discusses the distribution curves, cumulative frequency curves, special representations, and logarithmic class intervals. For large samples of a continuously varying property, rather small class intervals can be selected. In the limiting situation, infinitesimally small class widths and a sample of infinite size, the histogram of the relative frequencies becomes the distribution curve of the numerical variable. A distribution curve will be either asymmetrical or symmetrical in regard to some symmetry axis. The curve may have one or more maxima and they may be located anywhere. In the geosciences, many distribution curves have only one peak, so that the occurrence of two maxima in a sample distribution may indicate that two different populations with different maxima were blended in the past by some geological or biological process. It is not known exactly what the distribution curve of an empirical variable is because all samples are of finite size.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call