Ttykuu

In order to analyze simulated data, @risk provides a histogram with handle bars, so that we can move these handles to see what proportion of data are on each side. For example,

.

I am trying to create similar graph for my empirical data, but I have no idea how to do it. Any help starting with the following code would be much appreciated.

 mydata = RandomVariate[TriangularDistribution[100, 400, 200], 5000];
 Histogram[mydata, Automatic, "Probability"]
 mydist = EmpiricalDistribution[mydata];

Try this:

With[dist = CDF[mydist],
 Manipulate[
 Histogram[mydata, Automatic, "Probability",
 PlotRange -> MinMax@mydata, All,
 PlotLabel -> 
 StringForm["Percent Inside: ``" , 
 dist[bounds[[2]]] - dist[bounds[[1]]]],
 Epilog ->
 
 EdgeForm[None],
 GrayLevel[1, .5],
 Rectangle[
 Min@mydata, 0,
 bounds[[1]], 1000000
 ],
 Rectangle[
 bounds[[2]], 0,
 Max@mydata, 1000000
 ]
 
 ],
 
 bounds, Rescale[.25, .75, 0, 1, MinMax@mydata],
 Sequence @@ MinMax@mydata,
 ControlType -> IntervalSlider
 
 ]
 ]

A starting point:

BlockRandom[SeedRandom[42]; (* for reproducibility *)
 mydata = RandomVariate[TriangularDistribution[100, 400, 200], 5000]];

DynamicModule[hd, hist, xmin, xmax, yr,
 Manipulate[Show[hist, 
 Plot[PDF[hd, x], x, h[[1, 1]], h[[2, 1]],
 Exclusions -> None, Filling -> 1 -> Axis,
 FillingStyle -> Opacity[0.8, Pink],
 PlotPoints -> 95, PlotRange -> yr, 
 PlotStyle -> Directive[AbsoluteThickness[2], Pink]],
 Axes -> None, Frame -> True,
 GridLines -> h[[1, 1]], h[[2, 1]], None, 
 Method -> "GridLinesInFront" -> True, 
 PlotLabel -> StringForm["``%",
 100 Round[Abs[CDF[hd, h[[2, 1]]] -
 CDF[hd, h[[1, 1]]]],
 0.001]]],
 h, xmin, 0, xmax, 0, Locator, Appearance -> None,
 Initialization :> (hd = HistogramDistribution[mydata];
 xmin, xmax = First[hd["Domain"]]; 
 hist = Histogram[mydata, Automatic, "PDF", 
 ChartBaseStyle ->
 Directive[EdgeForm[], LightPink]];
 yr = Last[Charting`get2DPlotRange[hist]];)]]