Correlation analysis between variables
Analyze correlations between dataset variables with business interpretation and heatmap visualization.
Paste in your AI
Paste this prompt in ChatGPT, Claude or Gemini and customize the variables in brackets.
J'ai un dataset contenant [NOMBRE_VARIABLES] variables : [LISTE_VARIABLES]. Le dataset comporte [NOMBRE_OBSERVATIONS] observations sur la période [PERIODE]. L'objectif est de comprendre les relations entre ces variables pour [OBJECTIF_METIER].
Réalise une analyse de corrélation complète :
- Calcule la matrice de corrélation (Pearson pour variables continues, Spearman si distribution non-normale, Kendall pour variables ordinales)
- Identifie les paires de variables avec corrélation forte (|r| > 0.7) et modérée (|r| > 0.4)
- Distingue corrélation et causalité pour les relations identifiées
- Détecte la multicolinéarité problématique pour de futures analyses de régression
- Représente graphiquement la matrice de corrélation (heatmap) avec le code approprié
- Propose des hypothèses métier pour expliquer les corrélations les plus significatives
- Identifie des variables potentiellement redondantes pouvant être supprimées
Contexte business : [CONTEXTE_BUSINESS]
Outil préféré : [OUTIL] (Python pandas/seaborn, R, Excel)
Fournis une interprétation en langage accessible pour des non-statisticiens.
Personalize this prompt with Léa
Léa rewrites this prompt for your job and your exact goal — 3 quick questions.
Why this prompt works
<p>The prompt is effective because it requests both technical calculation and accessible business interpretation, bridging the gap between statistical analysis and managerial decision-making. It avoids the correlation/causation trap.</p>
Use Cases
Expected Output
Annotated correlation matrix, identification of significant relationships, visualization code, and business interpretations.
Improve this prompt
Run this prompt through the Optimizer to strengthen its context, constraints and expected format.
Improve this prompt with the OptimizerComments
- LéaAI
Pour détecter des relations non linéaires ignorées par Pearson, ajoutez une étape de visualisation par paires (scatter matrix) sur les variables à forte corrélation. Cela évite de fausses interprétations et affine vos hypothèses métier.
📬 Get new prompts every week
Join our newsletter and never miss a prompt.
Go further
Similar Prompts
Stock and inventory data analysis
Complete inventory data analysis with ABC/XYZ classification, threshold calculations, and working capital optimization.
Retention Analysis
Improve user retention
Normalize features for deep learning
Prepare data for deep learning
DALL-E Prompt for Analyzing Market Trends
DALL-E, OpenAI's image generation model, offers a unique visual approach to exploring and communicating market trends. While DALL-E is not a traditional data analysis tool, it can create impactful visual representations of market dynamics: conceptual infographics, stylized data visualizations, forward-looking scenario illustrations, or mood boards reflecting shifting consumer preferences. By combining precise instructions about identified trends with DALL-E's creative capabilities, analysts and marketers can produce visual aids that make market insights instantly understandable for teams and stakeholders. This approach is especially effective for strategic presentations, competitive intelligence reports, and foresight documents where visual impact enhances data comprehension. Discover how to craft your DALL-E prompts to turn your trend analysis into professional, memorable visuals.