Run a Regression Analysis
Set up and interpret a regression analysis to understand what factors drive a business outcome.
The Prompt
Help me run a regression analysis on the following data. Provide: 1. Model selection: linear / logistic / multiple / polynomial — with reasoning 2. Variable selection: which to include, interactions to consider 3. Python/R code to run the analysis 4. How to interpret the coefficients 5. Diagnostic checks: multicollinearity, residual plots, outlier influence 6. How to present results to non-technical stakeholders 7. Limitations and caveats Research question: [WHAT ARE YOU TRYING TO PREDICT OR EXPLAIN?] Outcome variable: [WHAT YOU'RE PREDICTING] Predictor variables: [LIST YOUR INDEPENDENT VARIABLES] Data size: [NUMBER OF OBSERVATIONS] Language: [PYTHON (sklearn/statsmodels) / R]
Example Output
Multiple linear regression to predict 30-day LTV from signup characteristics. Code using statsmodels OLS with company_size, plan_at_signup, and days_to_first_api_call as predictors. Results: days_to_first_api_call has the strongest negative coefficient (-$45 LTV per day delayed). R² = 0.61. VIF scores all <5 (no multicollinearity).
FAQ
Which AI model is best for Run a Regression Analysis?
Claude Sonnet 4 — strong at statistics with clear plain-English interpretation.
How do I use the Run a Regression Analysis prompt?
Copy the prompt, replace the [BRACKETED] placeholders with your specific information, and paste into your preferred AI assistant (ChatGPT, Claude, Gemini, etc.). Multiple linear regression to predict 30-day LTV from signup characteristics. Code using statsmodels OLS with company_size, plan_at_signup, and days_to_first_api_call as predictors. Results: days_to_first_api_call has the strongest negative coefficient (-$45 LTV per day delayed). R² = 0.61. VIF scores all <5 (no multicollinearity).
Model Recommendation
Claude Sonnet 4 — strong at statistics with clear plain-English interpretation.