The objective of the macro is to pass the 3 tests of fit based on Empirical Distribution Function (EDF) statistics
(Kolmogorov-Smirnov (D), Cramer-VonMises (W2)
and Anderson-Darling (A2)) on the attribute "E" in the
"youngsmodulus" dataset. The tested law is the "normal"
distribution when both the mean (30576) and variance (1450) are set or when both are defined either from the sample.
"""
Example of distribution testing with quality criteria on data
"""
from rootlogon import ROOT, DataServer, UncertModeler
tds = DataServer.TDataServer()
tds.fileDataRead("youngsmodulus.dat")
c = ROOT.TCanvas("c1", "Test on youngsmodulus dataset", 13, 38, 1210, 1874)
pad = ROOT.TPad("pad", "pad", 0, 0.03, 1, 1)
pad.Draw()
pad.Divide(1, 3)
tks = UncertModeler.TTestKolmogorovSmirnov(tds, "E")
pad.cd(1)
tks.computeScore("normal:normal(30576,1450)")
tcvm = UncertModeler.TTestCramerVonMises(tds, "E")
pad.cd(2)
tcvm.computeScore("normal:normal(30576,1450)")
tad = UncertModeler.TTestAndersonDarling(tds, "E")
pad.cd(3)
tad.computeScore("normal:normal(30576,1450)")
