Have you heard of the word distributions? I think, this word is most occurring words in the statistics. Let's try to understand the word 'Distribution' first.
Meaning of Distribution:
Distribution means how the random variable is spread across when we out all the random variables on a X and Y axis. Distribution is mainly classified into two categories, Population Distributions and Sample Distributions.
Classification of Distribution:
I have made a classification in the Microsoft PowerPoint. As you can see that, mainly it is classified in two major types. First is the Population Distribution and second is Sampling Distribution. If there is an event or out come of experiment that we perform. Then it is most likely that if we note the outcomes of the experiment, it will be similar to one of the distributions mentioned in the classification chart below.
As you can see mainly the distribution is divided into two categories. In this blog, I will explain the types of Population Distribution as it is most widely used. There are two types while dealing with Population. Population can be Discrete Random Variable (1,2,3,4,6,111,3344, etc.) and Continuous Random Variables (10.1,10.2,11.0,11.5,12.9 etc.). Based on what is the random variable, it is divided in two types. We can see below, explanation of each distribution in short.
Understanding Each Distribution:
1. Binomial Distribution: This distribution is plotted which has only two outcomes. If the event is tossing a coin where we will get Heads or Tail, Shipment Delayed or not delayed, Men or women, Good grades or bad grades, Defective or not defective, Stock price will increase or decrease, Pass or Fail, Positive or Negative etc. This distribution will give the chance of happening one of them.
Only two outcomes for each event.
Discrete distribution
Should know the sample size
Should know the probability of success
2. Poisson Distribution: This distribution gives the probability of what is the chance of occurring the event in some time frame or some rate. For example, if we have to find the customers coming in 1 hour interval in shop or any store, we have to follow a poisson distribution. This is because, Poisson Distribution is only used when you have a very rare occurrences.
It is a discrete distribution.
It describes rare events.
Each occurrence is independent of the other occurrences.
It describes discrete occurrences over a continuum or interval.
The occurrences in each interval can range from zero to infinity.
The expected number of occurrences must be constant throughout experiment.
3. Hypergeometric Distribution: Like Binomial Distribution, there are two outcomes Success and Failure for Hypergeometric Distribution. The only difference is that Hypergeometric Distribution is used when sampling is done without replacement. Also, we should know the population that we want to study, because we are replacing the samples that we are drawing from population. These three were Discrete distributions and now we will see types of Continuous distribution.
4. Uniform Distribution: This distribution comes under continuous distribution. In Uniform Distribution has all outcomes has same chance of occurring.
5. Normal Distribution: This was developed by Carl Friedrich Gauss. Thus, also called as Gaussian Distribution. Normal distribution is famous distribution as many times in real life and natural data is distributed Normally. This distribution has same mean and same median. This is Bell shaped curve when plotted and is symmetric around the mean.
6. Exponential Distribution: This distribution is concerned with the amount of time until some specific event occurs. For example, amount of time in years the laptop battery lasts, amount of time clerk spends time with customer in Bank. This distribution has 'm' variable also called as Rate of Change. 'm' is calculated as 1/mean. Mean is the average time spend with customers or average time laptop battery lasts.
7. T Distribution: T Distribution falls under continuous distribution. This distribution is used when we have less number of samples 'n' from the population. Rule of thumb is to use this distribution if the 'n' value is less than 30.
8. Chi Square Distribution: This distribution falls under Continuous Distribution. This distribution is used to find how two things are related, such as delay at two locations. If we want to find reasons for delay while delivering an Amazon package at two or more locations. We can answer this question using Hypothesis Test.
9. F Distribution: This distribution falls under Continuous Distribution.
Thank you for reading this blog. Keep learning and if you have any doubt or want any specific topic, please let me know. and I will try to mention in my blog.
Thanks,
Soham Sanjay Shinde