statsmodels

1# Statsmodels: Statistical Modeling and Econometrics
2 
3## Overview
4 
5Statsmodels is Python's premier library for statistical modeling, providing tools for estimation, inference, and diagnostics across a wide range of statistical methods. Apply this skill for rigorous statistical analysis, from simple linear regression to complex time series models and econometric analyses.
6 
7## When to Use This Skill
8 
9This skill should be used when:
10- Fitting regression models (OLS, WLS, GLS, quantile regression)
11- Performing generalized linear modeling (logistic, Poisson, Gamma, etc.)
12- Analyzing discrete outcomes (binary, multinomial, count, ordinal)
13- Conducting time series analysis (ARIMA, SARIMAX, VAR, forecasting)
14- Running statistical tests and diagnostics
15- Testing model assumptions (heteroskedasticity, autocorrelation, normality)
16- Detecting outliers and influential observations
17- Comparing models (AIC/BIC, likelihood ratio tests)
18- Estimating causal effects
19- Producing publication-ready statistical tables and inference
20 
21## Quick Start Guide
22 
23### Linear Regression (OLS)
24 
25```python
26import statsmodels.api as sm
27import numpy as np
28import pandas as pd
29 
30# Prepare data - ALWAYS add constant for intercept
31X = sm.add_constant(X_data)
32 
33# Fit OLS model
34model = sm.OLS(y, X)
35results = model.fit()
36 
37# View comprehensive results
38print(results.summary())
39 
40# Key results
41print(f"R-squared: {results.rsquared:.4f}")
42print(f"Coefficients:\\n{results.params}")
43print(f"P-values:\\n{results.pvalues}")
44…