Index

Using Histograms to Explore Data

All the examples which follow use the same set of values – the sum of three dice, thrown repeatedly. Clearly the range will vary from 3 to 18, with the more likely values falling in the middle.

Aggregating a Set of Samples

This first chart takes most of the SharpPlot defaults as they come and simply displays the distribution of samples.

sp.Heading = "Sum of 3 Dice Throws";
threedice = new int[] {9,8,6,11,9,14,11,8,9,14,12,7,6,11,15,14,11,7,5,8,
     7,9,6,7,12,7,13,8,9,13,7,7,8,12,11,16,12,9,11,12,12,12,9,14,8,9,13,
     11,10,13,11,13,9,10,10,4,14,8,9,11,12,5,12,15};

sp.HistogramStyle = HistogramStyles.SurfaceShading;
sp.SetXRange(3,18);
sp.XAxisStyle = XAxisStyles.MiddleLabels;

sp.DrawHistogram(threedice);

With integer data (one ‘bin’ per value) it clearly makes sense to label the bars, rather than the tickmarks. For some reason, thise set of throws produced very few 10s, but is otherwise unremarkable.

Grouping into Class Intervals

This example aggregates the same data into groups 2-4, 4-6 and so on. The groups are all labelled (including the groups with zero count, as this is perfectly valid data here).

sp.Heading = "3 Dice Throws (Grouped)";
sp.HistogramStyle = HistogramStyles.SurfaceShading|HistogramStyles.ValueTags;
sp.SetFillStyles(FillStyle.GradientBottom);
sp.SetValueFont("Arial",10,FontStyle.Bold);

sp.SetXRange(3,18);
sp.ClassInterval = 2;
sp.SetXTickMarks(2);

sp.DrawHistogram(threedice);

Very Plain, with Exploded Axes

This example is a style of histogram often seen in statistical journals.

sp.Heading = "Histogram with Exploded Axes";
sp.HistogramStyle = HistogramStyles.Risers|HistogramStyles.ExplodeAxes;

sp.SetXRange(3,18);
sp.ClassInterval = 2;
sp.SetXTickMarks(2);

sp.DrawHistogram(threedice);

Overlayed Normal Curve

The final example draws a theoretical bellcurve centred on the mean.

sp.Heading = "Histogram with Normal curve";
sp.HistogramStyle = HistogramStyles.SurfaceShading|HistogramStyles.Risers|HistogramStyles.NormalCurve;
sp.SetFillStyles(FillStyle.Halftone);

sp.SetPenWidths(2);
sp.SetXRange(3,18);
sp.ClassInterval = 3;
sp.SetXTickMarks(3);

sp.DrawHistogram(threedice);

Datumlines or an XZone could be added to show the ‘range’ of the data, typically the mean +/- 2 standard deviations.