Parameter Estimation from Quantity Data: New in Wolfram Language 11

Parameter Estimation from Quantity Data

Get the lengths of major rivers in North America in miles.

In[1]:=

data = Quantity[ExampleData[{"Statistics", "RiverLengths"}], "Miles"];

In[2]:=

h = Histogram[data, 30, PDF, Frame -> True, FrameLabel -> Automatic, 
  PlotTheme -> "Detailed"]

Out[2]=

EstimatedDistribution accepts quantity data and returns QuantityDistribution with the units of the data.

In[3]:=

edist1 = EstimatedDistribution[data, 
  InverseGammaDistribution[a, b, c, d]]

Out[3]=

Estimate model in different units.

In[4]:=

edist2 = EstimatedDistribution[data, 
  QuantityDistribution[InverseGammaDistribution[a, b, c, d], 
   "Kilometers"]]

Out[4]=

Convert the first model to kilometers.

In[5]:=

UnitConvert[edist1, "Kilometers"]

Out[5]=

Plot the data histogram together with the fitted model PDF.

In[6]:=

Show[h, Plot[PDF[edist1, Quantity[x, "Miles"]], {x, 0, 2000}]]

Out[6]=