For sample data, you should first decide if your data is continuous or discrete. Discrete distributions always return integer values. For example, presume you have a set of data describing the number of failures in a series of 100 trial batches. You would only want to fit discrete distributions to this set because partial failures are not allowed. In contrast, continuous data can take on any value in a range. For example, presume you have a set of data describing the height, in inches, of 300 people. You would want to fit continuous distributions to this data, since heights are not restricted to integral values.
If you specify that your data is discrete, all your data values must be integers. Keep in mind, however, that the converse is not true. Just because you have all integral data values does not mean you have to fit only discrete distributions. In the previous example, the height data set may be rounded to the nearest inch, but fitting to continuous distributions is still appropriate.
BestFit does not support the fitting of discrete distributions to density and cumulative curve data.
You can specify whether your data set is continuous or discrete in the Input Data Options dialog.
