Handling-Missing-Values

Missing data, also known as missing values, is where some of the observations in a data set are blank.

ex.

• At age feature womens hesitate to put down their age
• Men hesitate to show their salary
• informations are not that valid

Type of data

1. Categorical Data: 
    It is a string type of data such as Gender, Sex, education, etc.

2. Discrite Data:
    It is a number type data which is whole number only such as How many bank account you have,How many bike you have, etc.

3.Continous Data:
    It is a number type data such as Age, Height, Profit, etc.

Different type of Missing data

1. Missing Completely at Random (MCAR)
    - The variable is missing completely at random (MCAR) if the probability of being missing is the same for all the observations.
    - When data is MCAR, there is absolutely no relationship between the data missing and other values.

2. Missing Data Not at Random (MNAR)
    - There is absolutely some relationship between the data misssing and any other feature's  values in dataset. 

3. Missing at Random(MAR)
    - If the propensity for a data point to be missing is't related to the missing data, but it's related to some of the observed data.
    - When Data is MAR, The data is missing but can be predicted from other information.

All techniques of handling Missing values

1. Mean/Mode/Median replacement
2. Random sample imputation
3. Capturing NAN values with a new feature
4. End of Distribution imputation
5. Arbitrary imputation
6. Frequent categories imputation   

Note we perform all this techniques on 'Titanic' dataset, you can download it from :

• Titanic Datset