Scribd
Upload
22 views14 pages

Handling Missing Data

The document discusses strategies for handling missing data in datasets. It identifies different types of missing data like missing completely at random, missing at random, and missing not at random. It then describes common approaches to handle missing data like listwise deletion, mean imputation, predictive imputation, and forward fill. The document emphasizes choosing the right approach based on the type of missing data and performing sensitivity analysis. It also stresses the importance of reporting missing data and preventing missing data during data collection.

Uploaded by

Tom Cruise
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
22 views14 pages

Handling Missing Data

The document discusses strategies for handling missing data in datasets. It identifies different types of missing data like missing completely at random, missing at random, and missing not at random. It then describes common approaches to handle missing data like listwise deletion, mean imputation, predictive imputation, and forward fill. The document emphasizes choosing the right approach based on the type of missing data and performing sensitivity analysis. It also stresses the importance of reporting missing data and preventing missing data during data collection.

Uploaded by

Tom Cruise
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 14

Learn to Handle

Missing Data
Improve Data Analysis Quality by 100%
The Importance of
Addressing Missing Data

created by Bosscoder Academy


1
Identifying Missing Data

Example Dataset:

Missing Completely at Random (MCAR):

| ID | Age | Gender | Income |

|----|-----|--------|--------|

| 1 | 25 | M | 50000 |

| 2 | NaN | F | 60000 |

| 3 | 30 | M | NaN |

| 4 | 28 | F | 55000 |

| 5 | 35 | NaN | 70000 |

created by Bosscoder Academy


2
Types of Missing Data

created by Bosscoder Academy


3
Example Dataset:

Missing at Random (MAR):

| ID | Age | Gender | Income |

|----|-----|--------|--------|

| 1 | 25 | M | 50000 |

| 2 | NaN | F | 60000 |

| 3 | 30 | M | NaN |

| 4 | 28 | F | 55000 |

| 5 | NaN | M | 70000 |

Missing Not at Random (MNAR):

| ID | Age | Gender | Income |

|----|-----|--------|--------|

| 1 | 25 | M | 50000 |

| 2 | 28 | F | 60000 |

| 3 | 30 | M | 56250 |

| 4 | 28 | F | 55000 |

| 5 | 27 | M | NaN |

created by Bosscoder Academy


4
Strategies for Handling
Missing Data

Example Dataset:
after Listwise Deletion:
| ID | Age | Gender | Income |

|----|-----|--------|--------|

| 1 | 25 | M | 50000 |

| 4 | 28 | F | 55000 |

created by Bosscoder Academy


5
6
Example Dataset:
with Mean Imputation (MCAR):

| ID | Age | Gender | Income |

|----|-----|--------|--------|

| 1 | 25 | M | 50000 |

| 2 | 29.3| F | 60000 |

| 3 | 30 | M | 58500 |

| 4 | 28 | F | 55000 |

| 5 | 29.3| M | 70000 |

with Predictive Imputation (MAR):

| ID | Age | Gender | Income |

|----|-----|--------|--------|

| 1 | 25 | M | 50000 |

| 2 | 28 | F | 60000 |

| 3 | 30 | M | 56250 |

| 4 | 28 | F | 55000 |

| 5 | 27 | M | 72500 |

with Forward Fill (MNAR):

| ID | Age | Gender | Income |

|----|-----|--------|--------|

| 1 | 25 | M | 50000 |

| 2 | 25 | F | 60000 |

| 3 | 30 | M | 60000 |

| 4 | 28 | F | 55000 |

| 5 | 27 | M | 55000 |

7
Choosing the Right
Approach

created by Bosscoder Academy


8
Sensitivity Analysis

created by Bosscoder Academy


9
Reporting Missing Data

created by Bosscoder Academy


10
Prevention is Better Than
Cure

created by Bosscoder Academy


11
Conclusion

created by Bosscoder Academy


12
Why

Bosscoder?
750+ Alumni placed at Top
Product-based companies.

More than 136% hike for every 



2 out of 3 working professional.

Average package of 24LPA.

Explore More

You might also like