Handle Missing Data with Imputation
Data Preparation & Feature Engineering DS practice problem on Onlearn.
Difficulty: medium.
Topics: Understanding Data Imputation Strategies for Numerical and Categorical Features, Null value detection, Central tendency estimation, Categorical encoding, Imputer object instantiation, Transforming feature spaces, Data Preprocessing, Statistical Analysis, Exploratory Data Analysis, Feature Engineering, Data Cleaning, Missing Data Mechanisms (MCAR, MAR, MNAR), Univariate Imputation, Multivariate Imputation, Data Type Handling, Pandas Data Manipulation.
Implement a function that takes a pandas DataFrame and a strategy ('mean', 'median', or 'most frequent') to impute missing numerical values. If the strategy is 'mean' or 'median', apply it to numerical columns; if 'most frequent', apply it to all columns. Ensure the function returns the cleaned DataFrame.