Make Interaction column optional in fit_logit (#164)

* Initial plan

* Make Interaction column optional in fit_logit and add documentation

Co-authored-by: crvernon <3947069+crvernon@users.noreply.github.com>

---------

Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: crvernon <3947069+crvernon@users.noreply.github.com>

Author: Copilot

msdbook/tests/test_utils.py

@@ -179,4 +179,31 @@ def test_fit_logit_comprehensive(sample_data):
     assert np.all(np.isfinite(result.pvalues))  # P-values should be finite numbers
     # Check if any coefficient has a p-value less than 0.1 (10% significance level)
     assert np.any(result.pvalues < 0.1)
+
+
+def test_fit_logit_without_interaction():
+    """Test that fit_logit works without an Interaction column."""
+    np.random.seed(42)
+    n = 100
+    
+    # Create data WITHOUT an Interaction column
+    df_no_interaction = pd.DataFrame({
+        'Success': np.random.randint(0, 2, size=n),
+        'Predictor1': np.random.randn(n),
+        'Predictor2': np.random.randn(n)
+    })
+    
+    # This should work without raising a KeyError
+    result = fit_logit(df_no_interaction, ['Predictor1', 'Predictor2'])
+    
+    # Verify the result is valid
+    assert result is not None
+    assert hasattr(result, 'params')
+    
+    # Should have 3 parameters: Intercept, Predictor1, Predictor2 (no Interaction)
+    assert len(result.params) == 3
+    assert 'Intercept' in result.params.index
+    assert 'Predictor1' in result.params.index
+    assert 'Predictor2' in result.params.index
+    assert 'Interaction' not in result.params.index
 

msdbook/utils.py

@@ -4,13 +4,37 @@
 import statsmodels.api as sm
 
 def fit_logit(dta, predictors):
-    """Logistic regression"""
+    """Logistic regression with optional interaction term.
+    
+    Parameters
+    ----------
+    dta : pandas.DataFrame
+        Input data containing the 'Success' column and predictor columns.
+        If an 'Interaction' column is present, it will be included in the model.
+    predictors : list of str
+        List of predictor column names to include in the model.
+    
+    Returns
+    -------
+    statsmodels.discrete.discrete_model.BinaryResultsWrapper
+        Fitted logistic regression model.
+    
+    Notes
+    -----
+    The function automatically adds an intercept column. If the data contains
+    an 'Interaction' column, it will be included in the model alongside the
+    specified predictors.
+    """
 
     # Add intercept column of 1s
     dta["Intercept"] = np.ones(np.shape(dta)[0])
 
-    # Get columns of predictors
-    cols = dta.columns.tolist()[-1:] + predictors + ["Interaction"]
+    # Get columns of predictors, starting with intercept
+    cols = dta.columns.tolist()[-1:] + predictors
+    
+    # Add interaction term if present in the data
+    if "Interaction" in dta.columns:
+        cols = cols + ["Interaction"]
 
     # Fit logistic regression without the deprecated 'disp' argument
     logit = sm.Logit(dta["Success"], dta[cols])