Artificially introduced MAR dataset for missing data analysis in activity recognition

This dataset includes Missing at Random (MAR) data artificially introduced into activity recognition measurements gathered from a smart bracelet. The dataset is structured to support studies on the effectiveness of different imputation methods when data is missing in a non-random manner. It comprises data collected during activities such as walking, running, climbing stairs, and resting, with missing values selectively introduced under MAR conditions, making it suitable for testing realistic data recovery scenarios in wearable sensor data applications.