Machine learning vs. traditional logistic regression: predictive performance and risk factor identification for child nutritional outcome in Pakistan
Bibliographic record
Abstract
Logistic regression (LR) has long been the standard econometric tool for modeling child nutritional outcomes in public health research. However, conventional econometric LR (CE-LR) faces limitations in predictive accuracy, reliance on restrictive assumptions, and handling high-dimensional data. Machine learning-enhanced LR (ML-LR)-which relaxes the strict statistical assumptions of traditional models to better capture complex patterns-combined with Shapley Additive Explanations (SHAP), offers a promising alternative, improving both prediction and interpretability of risk factors. This study presents the first Pakistan-specific application and comparison of ML-LR (with SHAP analysis) against CE-LR, introducing a novel hybrid framework that combines predictive power with interpretability for policy-relevant insights using nationally representative data from Pakistan's 2017-2018 Demographic and Health Survey (n = 4,098 children under five). Results indicate persistent malnutrition rates: stunting (38.13%), underweight (23.04%), and wasting (8.05%). The ML-LR model identified all 13 hypothesized risk factors as significant, while CE-LR detected only six. Crucially, ML-LR captured key predictors missed by CE-LR, such as maternal BMI, employment, and dietary diversity. The SHAP analysis further revealed nuanced relationships: child age, low maternal BMI, unemployment, and unimproved water increased malnutrition risk, while higher birth order, adequate dietary diversity (if children were given ≥ 5 food items), maternal education, and male gender had protective effects. Crucially, ML-LR + SHAP uncovered context-dependent relationships invisible to CE-LR. For example, dietary diversity operated bidirectionally-low diversity was a risk factor, while adequate diversity was protective-a distinction CE-LR failed to capture. These findings demonstrate ML-LR's superior ability to model complex, heterogeneous determinants of child malnutrition. The study advocates for integrating ML techniques with explainable AI (e.g., SHAP) in econometric analyses to enhance policy-relevant insights in public health.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.009 | 0.026 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.002 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".