Statistical Distribution Functions

SDF_Title

This post is mainly about observing the theoretical distributions with variation of the terms that make the equations. There will be a lot of animations which will hopefully help improve your understanding of these concepts.

  1. Normal Distribution
  2. Exponential Distribution
  3. Gamma Distribution
  4. Weibull Distribution

ND_Title

One of the most commonly seen distributions and one of the first ones taught in probability courses. We see this normal everywhere. Let us start with the expression.

\frac{e^{-\frac{(x-\mu )^2}{2 \sigma ^2}}}{\sqrt{2 \pi } \sigma }

We will now change the quantities one by one and see the variation in the graph. The one below is an example graph done by using \sigma = 1 and \mu = 1.

Standard_Normal

.

The graph below is the one with changing values of \mu or mean.

Normal_mu

.

The graph below is the one with changing values of \sigma or standard deviation. Standard deviation tells about the variability of a particular normal distribution. We can observe that the as the \sigma is increasing, the graph becomes shorter and wider. Also note that the \sigma does not change the position of the peak or the mean.

Normal_sigma


EXP_Title

The exponential distribution can be written as \lambda e^{\lambda x}

In the case of this distribution, the parameter of interest is \lambda and the mean value of the distribution is \frac{1}{\lambda }. The graph below is plotted for \lambda = 1.

This graph below will show what happens to the distribution with various values of \lambda

Exponential_Static

Exponential_Lambda


Gamma_Title

The gamma distribution can be expressed as  \frac{\beta ^{\alpha }}{\Gamma (\alpha)}  x^{\alpha -1} e^{-\beta x} where \alpha and \beta is the random variables and x is the random variable.

Let us look at some plots of the variable by changing \alpha and \beta independently.

 

\beta = 1; various values of \alpha

Gamma_Beta=1

\beta ranging from 0.01 to 1 in steps of 0.01 for \alpha values of 1 to 10.

Gamma_Alpha=1


Weibull_Title

Weibull Distribution is denoted by \frac{\alpha}{\beta^\alpha} x^{\alpha-1} e^{{-(\frac{x}{\beta}})^{\alpha}} where x is the random variable and  \alpha and \beta are the parameters. Let us look at the behavior of the distribution with various values of \alpha and \beta.

\alpha varying from 1 to 10 and \beta =1

Weibull_alpha_var

Weibull distribution with \alpha  from 1 to 10 and \beta changing from 1 to 10

Weibull_beta_var


Lognormal_Title

The lognormal distribution graphs will be added soon.

 

 

 

 

 

 

.