Handling Missing Data | Part 1 | Complete Case Analysis

Handling missing data is an essential step in the data preprocessing pipeline, ensuring that ML models are trained on high-quality, representative datasets, leading to more accurate and reliable predictions Techniques like imputation, dropping missing values, or advanced methods such as Multiple Imputation can be employed based on the nature and impact of missing data. Choosing the right strategy ensures the reliability and accuracy of your models.

============================
Do you want to learn from me?
============================

📱 Grow with us:

⌚Time Stamps⌚

00:00 - Intro
00:58 - Handling Missing Data
05:50 - Complete Case Analysis [CCA]
07:09 - Assumption for CCA
09:38 - Advantages and Disadvantages of CCA
11:39 - When to use CCA?
13:24 - Code Example

[ajaykushwaha-jemw]

You are the fist youtuber on youtube with zero dislike. It makes me happy.
Sir app ka effor kabile tareef hai !

[akash.deblanq]

I can't comprehend how much I've learned from your videos. Got my first silver medal in kaggle today. All credit goes to you.

Feature engineering is so important, I'm focusing really hard on all these topics and you've done an amazing job at making these thorough tutorials. You're a great teacher. 🙏

[MuhammadJunaid-yrjd]

You are the fist youtuber on youtube with zero dislike. It makes me happy.
Sir app ka effor kabile tareef hai

[GamerBoy-iijc]

Sir ap first hain jinho ne complete btaya k q or kb apply krna CCA wrna mostly har koi bs btaa deta k apply krna ye ni btata k q krna . Thank u so much Sir again for providing this knowledge.

[Sanjay_Singh_Bisht]

Real Guru. Dhnya ho gaya main, jabse aapki video dekhi hai.

[ayesha]

This was extremely helpful and exactly what I was looking for. Thank you

[Shahad]

This is Gold for new learners, Thanks Nitish

[AmbujRai-ftcx]

new_df= df.dropna(subset=cols) to drop the rows and keep the cols as it is i.e the new_df.shape= (17182, 13)

[yearsago]

During the code example, why have we removed cols where null value percentage of data is greater than 5%.

[jitendratrivedi]

Target to complete the playlist by 12th January 202. Deserve more views. You are doing a great job

[ajaykushwaha]

Guru ji, aap gajab ka padhate hain, maja aa jata hai.

[Aestheticdeeps]

thank you soo much sir, crystal clear😇

[GAMEZONEX]

Hi! Thank you for the wonderful playlist. I have a query can we remove missing values using XGBoost), or probabilistic methods like Bayesian statistics .

[SumitKumar-uqdg]

Hello sir. How to understand whether the missing data is missing at random or not

[shubhankarsharma]

after CCA we are left with 17000 rows of new__df and 19000 rows of df .how to concatenate them for modeelling

[niranjania]

cols= [var for var in df.columns if and (df[var].isnull().mean>0))

[gautamdinga]

what if % of missing values of an attribute is exactly 5% then ? should we perform CCA

[rachanakotha]

@CampusX If factual data is missing like manufacture year of a vehicle. Is it fine to impute it? (Size of ds: 20k)

[nikitha]

How to add this cca data back to main dataframe???

[niranjania]

TypeError: '>' not supported be tween instances of 'method' and 'int'