Statistical Modeling Guide

# Statistical Modeling Guide

Provide a comprehensive guide to statistical modeling for [ANALYST_TYPE] (e.g., data science student, business analyst, academic researcher, product manager) working with [DATA_TYPE] data to answer [RESEARCH_QUESTION].

## Context
- **Tools available:** [TOOLS] (e.g., Python/pandas/scikit-learn, R, Excel, SPSS, Stata)
- **Modeling goal:** [GOAL] (description, prediction, causal inference, classification)
- **Data size:** [DATA_SIZE] (rows and columns)
- **Dependent variable type:** [DV_TYPE] (continuous, binary, count, categorical, time series)

## Statistical Modeling Framework

### 1. Model Selection Guide
Create a decision tree for choosing the right model based on: research question type, dependent variable type, data structure, and assumptions. Cover: linear regression, logistic regression, decision trees, random forest, time series (ARIMA), and clustering.

### 2. Data Preparation
Describe the essential pre-modeling data preparation steps:
- Missing data handling (imputation strategies)
- Outlier detection and treatment
- Feature engineering and transformation
- Train/validation/test split strategy
- Handling class imbalance (for classification)

### 3. Model Building Process
Walk through the end-to-end modeling process for [RESEARCH_QUESTION]: exploratory data analysis → feature selection → baseline model → iterative improvement → final model.

### 4. Model Validation & Evaluation
Explain key evaluation metrics for this model type: RMSE/MAE (regression), accuracy/AUC/F1 (classification), and how to use cross-validation properly.

### 5. Interpreting & Communicating Results
Explain how to interpret model coefficients, feature importance, and confidence intervals — and how to present findings to a non-technical audience.

### 6. Common Mistakes
List the 8 most common statistical modeling errors and how to avoid them.