{
  "task_id": "F6-conf-01",
  "n_missing_rows": 215,
  "n_train": 2500,
  "n_test": 2501,
  "anova_f_stat": 0.263465716807118,
  "anova_p_value": 0.9015060507654888,
  "effect_estimate": 0.050170098401698796,
  "effect_ci_low": -0.31933577569409977,
  "effect_ci_high": 0.41967597249749733,
  "baseline_metric": 4.875704154642544,
  "model_metric": 4.883841522290305,
  "conclusion": "no_detectable_effect",
  "feasibility": "complete",
  "feasibility_reason": null,
  "data_issues": [
    {
      "issue": "215 rows with empty DGS10 field (US market holidays / no publication)",
      "handling": "Dropped before differencing, as prescribed; changes computed between consecutive available observations."
    },
    {
      "issue": "214 changes (excluding ordinary Fri->Mon weekends) span a holiday gap and thus cover more than one business day",
      "handling": "Kept and attributed to the weekday of t, as prescribed by TASK.md; noted as a limitation (inflates variance, mostly for Tuesdays after Monday holidays)."
    }
  ],
  "n_input_rows": {
    "DGS10.csv": 5217
  },
  "n_analysed": 5001,
  "random_seed": null,
  "supplementary": {
    "monday_contrast_method": "Welch two-sample t interval (unequal variances)",
    "welch_se_bp": 0.18837955539096196,
    "welch_df": 1548.9581965107577,
    "welch_p_value": 0.7900247217974709,
    "train_weekday_means_bp": {
      "Monday": -0.05434782608695652,
      "Tuesday": -0.40545808966861596,
      "Wednesday": -0.2054263565891473,
      "Thursday": 0.01383399209486166,
      "Friday": 0.11683168316831684
    },
    "rmse_train_unconditional_mean_bp": 4.8751986528852305,
    "dm_stat_model_minus_zero_sqloss": 1.8573748978469748,
    "dm_p_value": 0.06325783524850241,
    "levene_p_value": 1.0122651168712971e-07,
    "kruskal_p_value": 0.5176806169324906
  }
}