A fitted distribution line is a theoretical distribution curve calculated using parameter estimates derived from a sample or from historical values that you enter. Use fitted distribution lines to determine how well sample data follow a specific distribution. These distribution lines are usually overlaid with the actual data so that you can directly compare the empirical data to the hypothesized distribution.
The following graphs include a fitted normal distribution line. Each graph table displays the parameter estimates used to generate the line or curve.
A normal fitted distribution line appears by default, for example, on probability plots and empirical CDF plots. You can specify a different distribution for these graphs, or you can add a fitted distribution line to other graphs, such as histograms, when you create a graph.
On the Data Display tab for probability plots only, you can also specify the confidence level for the confidence interval. By default, confidence intervals show the 95% confidence bounds for the individual percentiles of the distribution. These intervals form the outer solid lines on the plot, and can be used to assess the precision of the individual percentile estimates. The confidence intervals should not be used to assess the distribution fit.
Hold the mouse pointer on the fitted distribution line to view a table of estimated percentiles. To copy the tooltip text, select the element that has the tooltip, then right-click and choose .