March 17, 2024

Standard Normal (Z) Distribution

\(\text{A random variable, }X\text{, with this pdf is said to have the Normal Distribution:}\)

\(f(x) = \frac{1}{\sqrt{2 \pi \sigma}} \cdot e^{\frac{-{(x-\mu)}^2}{2{\sigma}^2}}; \hspace{1 cm} -{\infty} < x < \infty\)

\(\text{The Normal pdf is symmetrical and has }E(X) = \mu\text{ and }V(X) = {\sigma}^2\)

\(\text{A normal random variable with }E(X) = \mu = 0\text{ and }V(X) = {\sigma}^2 = 1\text{ is called a Standard}\)
\(\text{Normal random variable.}\)

Standard Normal Distribution Plot

Chi-Square Distribution

\(\text{A Chi-Square random variable, }\chi^2 \text{ with }n-1 \text{ degrees of freedom, df is:}\)

\({\chi^2}_{n-1} = \frac{(n-1)S^2}{\sigma^2}\)
\(\text{where }S^2 \text{ is the sample variance and }\sigma^2 \text{ is the population variance.}\)

\(\text{The domain of Chi-Square pdf}:\hspace{0.3 cm}\chi^2>0\)

\(\text{Thus, the } \chi^2 \text{ distribution is NOT SYMMETRIC.}\)

Chi-Square Distribution plot

Standard Normal Distribution 3D Plot

Normal Random Variable Standardized to Standard Normal Random Variable

x <- seq(-10, 10, length.out = 100)
y <- seq(-10, 10, length.out = 100)

mean_x <- mean(x)
sd_x <- sd(x)

mean_y <- mean(y)
sd_y <- sd(y)

x <- (x - mean_x) / sd_x
#mean(x)
#sd(x)

y <- (y - mean_y) / sd_y
#mean(y)
#sd(y)

z <- outer(x, y, function(x, y) dnorm(sqrt(x^2 + y^2)))

xax <- list(title = "X",
            titlefont = list(family = "Modern Computer Roman"))
yax <- list(title = "Y",
            titlefont = list(family = "Modern Computer Roman"))
zax <- list(title = "Density",
            titlefont = list(family = "Modern Computer Roman"))

plot_ly(x = x, y = y, z = z, type = "surface", colorscale = "Rainbow") %>%
  layout(
    title = "Normal Distribution Standardized Plot",
    scene = list(xaxis = xax,
                 yaxis = yax,
                 zaxis = zax
    )
  )

Summary

As seen on the previous slide, after standardizing the random normal
distribution values to standard normal distribution values using the
formula:

\(Z = \frac{X - \mu}{\sigma}\)

the mean of x evaluated to: -4.4408921^{-18} and the standard
deviation of x evaluated to: 1. Similarly, the mean of y evaluated to: -4.4408921^{-18} and the standard deviation of y evaluated to: 1.
-4.4408921^{-18} is effectively 0 without any rounding errors.

This proves that a random variable \(X\) from a standard normal
distribution has \(\mu_x: 0\) and \(\sigma_x: 1\).

Thank You!