1[2 {"id":"C1","text":"Using only end-of-year-t information, the OLS coefficient on leverage is 0.264 (firm-clustered SE 0.041): 10 pp higher leverage is associated with ~2.6 pp higher next-year default probability.","results_keys":["beta_leverage","se_leverage_cluster"],"status":"supported","caveat":"Association in simulated data; LPM; default treated as repeatable event."},3 {"id":"C2","text":"The estimation sample has 1,800 firm-years from 300 firms after removing 40 exact duplicate rows and 300 firm-years (all origin 2022) with unobserved outcomes.","results_keys":["n_analysed","n_firms","n_input_rows","data_issues"],"status":"supported","caveat":""},4 {"id":"C3","text":"rating_2025 was excluded because it is a 2025 field built from realised defaults and sample-average leverage (look-ahead leakage).","results_keys":["variables_excluded","predictors_used"],"status":"supported","caveat":""},5 {"id":"C4","text":"The baseline next-year default rate is 12.1% and the model explains little variance (R2 = 0.035).","results_keys":["default_rate","r2"],"status":"supported","caveat":""}6]7