FindingData
Statistics Introduction
Statistics Introduction
Nov 28, 2020
the first question that came in our mind is what is statistics and where we will use it in our data
Using statistics, we can gain deeper and more fine grained insights into how exactly our data is structured and based on that structure how we can optimally apply other data science techniques to get even more information.
- So, Statistics is a use of mathematics to perform technical analysis of data.
- Statistics used in reviewing the data, analyzing the data and draw conclusion from data means transforming the data into meaningful information.
- in case of statistics, we should know what the problem data is than transform the data accordingly.
- and on the other hand, when we talking about machine learning we should have Data without data, it is nothing.
| statistics | machine learning |
|---|---|
| problem -> Data | Data -> solution |
Types of Statistics
There are two types of statistics
- Descriptive Statistics
- Inferential Statistics (predictive statistics, prescriptive statistics)
let's understand them by some real world examples
| No | Descriptive | Predictive | Prescriptive |
|---|---|---|---|
| 1 | Describe where am i lossing the sales and why? | Look at the data and give me an idea what will happen if i change something ( Eg : if i will increase the price than will people buy my product or not? ) | afer all the research and analysis now tell me what should do? ( Eg : what should i do so that sale of my product increase by 10% ) |
| 2 | How many centuries virat kohli did in previous match? | How many centuries will virat kholi do in the next match? | How do we get virat kohli score more than 10 centuries in the world cup? |
Descriptive Statistics
- The type of statistics dealing with numbers and figures or information to describe any phenomena.
- These numbers are descriptive statistics.
- you are going to understand your data by description statistics.
Examples:
- Reports of industry production- cricket batting averages- Moving ratings- etc ...Inferential Statistics
- Inferential statistics is a decision, estimation, or generalization of population based on sample.
If i do this --> what will happen?
population
population is a collection of all the individual or orbjects.
sample
Sample is subset of population.
examples
Example
Suppose in your college there are 1000 students. You are interested in finding out the how many students prefer eating in the college canteen than college mess.
A random group of 100 students is selected. Here our population size is of 1000 students and the sample size is of 100 students. You surveyed the sample group and got the following results:
| 1st year | 2nd Year | 3rd year | 4th Year | Total | |
|---|---|---|---|---|---|
| Canteen | 7 | 13 | 20 | 32 | 72 |
| Mess | 12 | 8 | 5 | 3 | 28 |
Look at the Total
- Descriptive
- 72 % of the students prefer eating in the canteen.
- 28 % of the students prefer eating in the mess.
- 1st year students are more inclined towards eating in mess.
- Inferential
- 3rd and 4th year students are the main target for sales of restaurant.
- You can give discounts to the 1st year students to increase the number count.
Inferential statistics is used to make inferences from data whereas description statistics simply describe what is going on in our data.
Next
Types of Data / Level of Measurement >
Edit this page on GitHub