Longitudinal Data Format¶

This tutorial introduces the data formats supported by Scikit-Longitudinal (Sklong), focusing on the wide format for longitudinal data. We'll explain why wide format is preferred to avoid data leakage and provide examples using a simple synthetic dataset.

Wide vs. Long Format¶

The animation below contrasts the two layouts at split time: a long-format split can cut through a single subject's rows and leak wave-level information across train/test, while a wide-format split always keeps one subject on one side.

Longitudinal data can be represented in two main formats:

• Long Format: Each observation for a subject is in a separate row, with a time indicator column. This can lead to data leakage during splitting if rows for the same subject are separated.

• Wide Format: Each subject is in one row, with columns for each feature at each time point (e.g., smoke_w1, smoke_w2). This format prevents leakage as all temporal data for a subject stays together.

Sklong focuses on wide format for safety and simplicity in machine learning workflows.

Synthetic Dataset Example¶

The dataset used is synthetic, mimicking health data for illustration.

• Longitudinal features: smoke, cholesterol, blood_pressure, diabetes, exercise, obesity (two waves each).
• Non-longitudinal: age, gender.
• Target: stroke_w2 (binary).

Load and prepare:

from scikit_longitudinal.data_preparation import LongitudinalDataset

dataset = LongitudinalDataset('./extended_stroke_longitudinal.csv')
dataset.load_data()
dataset.load_target(target_column='stroke_w2')
dataset.setup_features_group([[2,3], [4,5], [6,7], [8,9], [10,11], [12,13]]) # Indices for longitudinal features
dataset.load_train_test_split(test_size=0.2, random_state=42)

print(dataset.data.head())

This wide format ensures safe splitting and temporal integrity.