Normal distribution

Here's how we get the normal probability distribution function.

Plot[PDF[NormalDistribution[0, 1], x], {x, -3, 3}]

[Graphics:../HTMLFiles/normal_22.gif]

⁃Graphics⁃

The following definition returns the area (Cumulative Distribution Function) of the "standard normal distribution", i.e. a Gaussian with μ=0 and σ=1.  These are the numbers in table D-4 in the Understanding Statistics text, which is the probability for a value at or below the given z.

prob[z_] := CDF[NormalDistribution[0, 1], z] ;

Plot[prob[z], {z, -3, 3}]

[Graphics:../HTMLFiles/normal_26.gif]

⁃Graphics⁃

RowBox[{prob, [, RowBox[{-, 2.58}], ]}]

0.00494002

Since this is in the z<0 range, its the same as the NormalPValue result.

RowBox[{NormalPValue, [, RowBox[{-, 2.58}], ]}]

RowBox[{OneSidedPValue, , 0.00494002}]

But on the positive side we have

RowBox[{prob, [, 2., ]}]

0.97725

RowBox[{NormalPValue, [, 2., ]}]

RowBox[{OneSidedPValue, , 0.0227501}]

1 - OneSidedPValue/.NormalPValue[2]

0.97725

To find the inverse function, Mathematica uses the Quantile property of a distribution, which returns the z value where the distribution accumulates a given total probability.  So to find out the z value such that the probability below that z is 0.9775, we do this.

quan[p_] := Quantile[NormalDistribution[0, 1], p]

RowBox[{quan, [, 0.97725, ]}]

2.

Plot[quan[p], {p, 0, 1}]

[Graphics:../HTMLFiles/normal_42.gif]

⁃Graphics⁃


Created by Mathematica  (October 2, 2003)