{
  "name": "statsmodels binary-logit input, prediction and reporting contracts",
  "description": "Original crosswalk compiled from the official statsmodels Logit, formula, LogitResults and Pitfalls references. It separates required inputs, intercept handling, missing-value behavior and output scales so the same research model can be preserved from fitting through reporting.",
  "page": "https://casrai.org/guides/logistic-regression-statsmodels",
  "publisher": "CASRAI",
  "license": "https://creativecommons.org/licenses/by/4.0/",
  "dateModified": "2026-11-14T17:52:40Z",
  "temporalCoverage": null,
  "spatialCoverage": null,
  "columns": [
    {
      "name": "Workflow component",
      "unit": "",
      "description": ""
    },
    {
      "name": "Documented contract",
      "unit": "",
      "description": ""
    },
    {
      "name": "Check or implication",
      "unit": "",
      "description": ""
    },
    {
      "name": "Source",
      "unit": "",
      "description": ""
    }
  ],
  "rows": [
    {
      "Workflow component": "Array interface",
      "Documented contract": "Logit takes endog and exog; no intercept is added automatically.",
      "Check or implication": "Add an intercept explicitly when the intended model includes one.",
      "Source": "Logit API reference"
    },
    {
      "Workflow component": "Formula interface",
      "Documented contract": "Formula models accept a formula and data; the formula default includes an intercept.",
      "Check or implication": "Remove the intercept only when that is the intended specification.",
      "Source": "Fitting models using R-style formulas"
    },
    {
      "Workflow component": "Missing observations",
      "Documented contract": "Logit missing options: none, drop, raise. Default none performs no NaN check.",
      "Check or implication": "Choose and record missing-value handling explicitly.",
      "Source": "Logit API reference"
    },
    {
      "Workflow component": "Coefficient intervals",
      "Documented contract": "conf_int constructs confidence intervals for fitted parameters.",
      "Check or implication": "Transform coefficient interval endpoints when reporting odds ratios.",
      "Source": "LogitResults API reference"
    },
    {
      "Workflow component": "Stored fittedvalues",
      "Documented contract": "LogitResults.fittedvalues is the linear predictor XB.",
      "Check or implication": "Do not label this attribute as fitted probabilities.",
      "Source": "LogitResults API reference"
    },
    {
      "Workflow component": "Convergence",
      "Documented contract": "Convergence criteria can misleadingly indicate success; separation can yield infinite or nonunique parameters.",
      "Check or implication": "Review warnings and the design; a convergence flag alone is insufficient.",
      "Source": "Pitfalls"
    },
    {
      "Workflow component": "Retained alternative fits",
      "Documented contract": "Result instances may depend on attributes of their model instance.",
      "Check or implication": "Use separate model instances when keeping different fits.",
      "Source": "Pitfalls"
    }
  ],
  "sources": [
    {
      "title": "Logit API reference",
      "url": "https://www.statsmodels.org/stable/generated/statsmodels.discrete.discrete_model.Logit.html",
      "accessed": "2026-11-14"
    },
    {
      "title": "Fitting models using R-style formulas",
      "url": "https://www.statsmodels.org/stable/example_formulas.html",
      "accessed": "2026-11-14"
    },
    {
      "title": "LogitResults API reference",
      "url": "https://www.statsmodels.org/stable/generated/statsmodels.discrete.discrete_model.LogitResults.html",
      "accessed": "2026-11-14"
    },
    {
      "title": "Pitfalls",
      "url": "https://www.statsmodels.org/stable/pitfalls.html",
      "accessed": "2026-11-14"
    }
  ]
}