A standard normal distribution is a normal distribution with mean zero and variance 1.
i.
Mean = 0 and standard deviation = 1
ii. Lower quartile
Lower quartile is the z value corresponding to the lower 25%.
From z-tables or =NORM.S.INV(0.25) excel function, the lower quartile is − 0.6745 -0.6745 − 0.6745
iii. Upper quartile
Lower quartile is the z value corresponding to the lower 75%.
From z-tables or =NORM.S.INV(0.75) excel function, the lower quartile is 0.6745 0.6745 0.6745
iv.Inter-quartile range
Inter-quartile range is the difference between upper and lower quartile.
I Q R = 0.6745 + 0.6745 = 1.349 IQR=0.6745+0.6745=1.349 I QR = 0.6745 + 0.6745 = 1.349
v. Mean deviation
Generally, mean deviation of a normal distribution is as follows,
E [ ∣ X − μ ∣ ] = ∫ − ∞ ∞ ∣ a − μ ∣ f X ( a ) d a \mathbf{E}\left[\left|X-\mu\right|\right] =\int_{-\infty}^{\infty}\left|a-\mu\right|f_{X}(a)da\\ E [ ∣ X − μ ∣ ] = ∫ − ∞ ∞ ∣ a − μ ∣ f X ( a ) d a
= ∫ − ∞ μ ( μ − a ) f X ( a ) d a + ∫ μ ∞ ( a − μ ) f X ( a ) d a =\int_{-\infty}^{\mu}\left(\mu-a\right)f_{X}(a)da+\int_{\mu}^{\infty}\left(a-\mu\right)f_{X}(a)da\\ = ∫ − ∞ μ ( μ − a ) f X ( a ) d a + ∫ μ ∞ ( a − μ ) f X ( a ) d a
= 1 2 ∫ μ ∞ ( a − μ ) f X ( a ) d a \overset{1}{=}2\int_{\mu}^{\infty}\left(a-\mu\right)f_{X}(a)da\\ = 1 2 ∫ μ ∞ ( a − μ ) f X ( a ) d a
= 2 ∫ μ ∞ a − μ σ 2 π e − ( a − μ σ 2 ) 2 d a =2\int_{\mu}^{\infty}\frac{a-\mu}{\sigma\sqrt{2\pi}}e^{-\left(\frac{a-\mu}{\sigma\sqrt{2}}\right)^{2}}da\\ = 2 ∫ μ ∞ σ 2 π a − μ e − ( σ 2 a − μ ) 2 d a
= 2 2 π ∫ 0 ∞ b e − b 2 σ 2 d b \overset{2}{=}\frac{2}{\sqrt{\pi}}\int_{0}^{\infty}be^{-b^{2}}\sigma\sqrt{2}db\\ = 2 π 2 ∫ 0 ∞ b e − b 2 σ 2 d b
= 2 2 π σ ∫ 0 ∞ b e − b 2 d b =2\sqrt{\frac{2}{\pi}}\sigma\int_{0}^{\infty}be^{-b^{2}}db\\ = 2 π 2 σ ∫ 0 ∞ b e − b 2 d b
= 2 2 π σ [ e − b 2 − 2 ] 0 ∞ =2\sqrt{\frac{2}{\pi}}\sigma\left[\frac{e^{-b^{2}}}{-2}\right]_{0}^{\infty}\\ = 2 π 2 σ [ − 2 e − b 2 ] 0 ∞
= 2 π σ [ e 0 − e − ∞ ] =\sqrt{\frac{2}{\pi}}\sigma\left[e^{0}-e^{-\infty}\right]\\ = π 2 σ [ e 0 − e − ∞ ]
= 2 π σ . =\sqrt{\frac{2}{\pi}}\sigma. = π 2 σ .
But since standard deviation is 1 for a standard normal distribution, mean deviation becomes 2 π \sqrt{\frac{2}{\pi}} π 2
Comments