风控模型开发验证报告

00 模型概要信息

项目	内容
模型名称	exp_20261010-145348-8f87
开发完成日期	2026-10-10
模型开发人员	（人工填写）
模型开发人员邮件地址	（人工填写）
模型验证人员	（人工填写）
模型验证人员邮件地址	（人工填写）
模型应用环节	preloan_approval
模型应用频率	（人工填写）
模型开发背景及业务用途描述	Local MVP full pipeline run.
模型适用产品/客户描述	（人工填写）
模型重开发原因（如有）	—
模型拟监控频率	（人工填写）
模型路线	LR + WOE 评分卡
目标变量	flag_30
实验 ID	ITE-mv21ge5f-aa73

待补充：模型开发/验证人员、应用频率、产品描述等字段需人工补填。

01 模型设计

建模数据样本窗口

项目	内容
切分方式	外部 data_flag 划分（TRAIN 内 6/4 随机开发/测试，VALI 跨时间，PSI 时点）
开发样本(Train)窗口	1970-01-01 ~ 1970-01-01
预留验证样本窗口	1970-01-01 ~ 1970-01-01
跨时间验证样本窗口	1970-01-01 ~ 1970-01-01
PSI时点样本窗口	1970-01-01 ~ 1970-01-01

模型好坏定义

项目	内容
目标变量	flag_30
坏样本(BAD=1)占比	22.06%
好样本(BAD=0)占比	—
不确定样本	本流程要求目标变量已预计算为 0/1 二值，不确定样本请在上游合并口径

样本排除规则

序号	排除规则	说明
1	ScoreCust2	主键/时间/口径字段，不入模
2	apply_dt	主键/时间/口径字段，不入模
3	flag_30 为空	排除 0 行

滚动率分析

待补充：滚动率分析（M0→M6+ 状态迁移矩阵）需要逾期状态口径数据，当前流程目标变量为上游已计算的二值标签，暂未生成。

待补充：表现窗口及坏账率分析、模型细分定义需结合业务口径人工补充。

02 数据流

开发时点 — 样本排除

类别	样本数
全部样本	30000
排除：目标变量为空	0
排除后样本	30000

注：按照01模型设计的排除规则统计；当前流程的排除项为目标变量缺失。

总体样本概况

数据集	样本量	好样本	坏样本	坏样本率	占比
全部样本	30000	23383	6617	22.06%	100.00%
开发样本	10801	8453	2348	21.74%	36.00%
测试样本	7199	5635	1564	21.73%	24.00%
训练+测试样本	18000	14088	3912	21.73%	60.00%
跨时间验证样本	6000	4624	1376	22.93%	20.00%
PSI时点样本	6000	4671	1329	22.15%	20.00%

开发/预留验证样本交叉统计

类别	总样本数	开发样本数	开发占比	预留验证样本数	预留验证占比
好	23383	8453	36.15%	5635	24.10%
坏	6617	2348	35.48%	1564	23.64%
合计	30000	10801	36.00%	7199	24.00%

细致变量分析（预分箱口径，开发样本）

变量	IV	缺失率	取值集中度	最大相关系数
risk_score 内部风险评分	0.1434	0.00%	0.91%	0.9986
risk_score_v2 内部风险评分副本	0.1390	0.00%	0.94%	0.9986
bh_a001 百融履约指数A001	0.1382	0.00%	0.15%	0.9298
ScoreCust2_v2 客户评分卡2副本	0.1321	0.00%	1.14%	0.9138
bh_a009 百融信用历史长度	0.1271	0.00%	1.73%	0.8070
cust_seg2 客群分层等级	0.0848	0.00%	30.86%	0.6662
avg_tnr 平均履约月数	0.0841	0.00%	1.92%	0.6947
query_cnt_3m 近3月机构查询次数	0.0732	0.00%	16.53%	0.5822
income_grade 收入等级	0.0632	0.00%	31.12%	0.5458
repay_amt_6m_grade 近6月还款金额等级	0.0607	0.00%	27.69%	0.5342
rare_field 稀疏字段(高缺失)	0.0330	96.32%	0.76%	0.0812
tenure_m 授信时长(月)	0.0217	0.00%	5.50%	0.3328
card_usage_ratio 信用卡额度使用率	0.0214	0.00%	1.69%	0.3613
bank_997_cnt_3m 近3月银行查询次数	0.0205	0.00%	34.62%	0.2757
edu_level 学历	0.0004	0.00%	42.13%	—
region_code 区域	0.0004	0.00%	20.63%	—
credit_no	0.0000	0.00%	0.01%	—
flag_special 特殊标记(高集中度)	0.0000	0.00%	95.41%	0.0380

注：共 18 个候选变量，按 IV 降序展示前 18 个；集中度 = 最大频次取值样本数 ÷ 非空样本数；相关性为皮尔逊（仅数值变量），完整数据见 variable_detail.json。

03 单变量分析（细分箱）

risk_score 内部风险评分 — IV 0.1272，10 箱

分组号	分栏	Total	Good	Bad	Bad Rate	WOE
1	(-inf, 492.5]	1279	1232	47	3.67%	0.2165
2	(492.5, 519.5]	1276	1182	94	7.37%	0.1696
3	(519.5, 535.5]	1032	914	118	11.43%	0.1000
4	(535.5, 546.5]	801	694	107	13.36%	0.0648
5	(546.5, 564.5]	1350	1144	206	15.26%	0.0779
6	(564.5, 587.5]	1704	1320	384	22.54%	-0.0112
7	(587.5, 600.5]	870	626	244	28.05%	-0.0507
8	(600.5, 616.5]	797	531	266	33.38%	-0.0859
9	(616.5, 643.5]	955	529	426	44.61%	-0.1917
10	(643.5, inf]	737	281	456	61.87%	-0.2638

ScoreCust2_v2 客户评分卡2副本 — IV 0.1262，10 箱

分组号	分栏	Total	Good	Bad	Bad Rate	WOE
1	(-inf, 507.5]	1649	1574	75	4.55%	0.2546
2	(507.5, 522.5]	1083	979	104	9.60%	0.1235
3	(522.5, 536.5]	1285	1138	147	11.44%	0.1205
4	(536.5, 544.5]	790	677	113	14.30%	0.0567
5	(544.5, 553.5]	959	776	183	19.08%	0.0237
6	(553.5, 565.5]	1232	965	267	21.67%	0.0007
7	(565.5, 583.5]	1611	1171	440	27.31%	-0.0737
8	(583.5, 593.5]	655	432	223	34.05%	-0.0766
9	(593.5, 606.5]	640	370	270	42.19%	-0.1231
10	(606.5, inf]	897	371	526	58.64%	-0.2861

bh_a001 百融履约指数A001 — IV 0.1128，10 箱

分组号	分栏	Total	Good	Bad	Bad Rate	WOE
1	(-inf, 223.6500015258789]	1032	994	38	3.68%	0.1794
2	(223.6500015258789, 254.3499984741211]	1349	1266	83	6.15%	0.1937
3	(254.3499984741211, 280.0500030517578]	1560	1381	179	11.47%	0.1408
4	(280.0500030517578, 290.84999084472656]	786	672	114	14.50%	0.0549
5	(290.84999084472656, 317.0500030517578]	1927	1560	367	19.05%	0.0421
6	(317.0500030517578, 332.75]	1088	818	270	24.82%	-0.0301
7	(332.75, 354.0500030517578]	1197	825	372	31.08%	-0.0969
8	(354.0500030517578, 367.75]	542	334	208	38.38%	-0.0871
9	(367.75, 394.25]	756	392	364	48.15%	-0.1814
10	(394.25, inf]	564	211	353	62.59%	-0.2142

cust_seg2 客群分层等级 — IV 0.1060，4 箱

分组号	分栏	Total	Good	Bad	Bad Rate	WOE
1	(-inf, 3.5]	2056	1867	189	9.19%	0.2166
2	(3.5, 4.5]	3333	2822	511	15.33%	0.1501
3	(4.5, 5.5]	3330	2505	825	24.77%	-0.0668
4	(5.5, inf]	2082	1259	823	39.53%	-0.2705

bh_a009 百融信用历史长度 — IV 0.0991，10 箱

分组号	分栏	Total	Good	Bad	Bad Rate	WOE
1	(-inf, 6.3500001430511475]	902	433	469	52.00%	-0.2386
2	(6.3500001430511475, 7.3500001430511475]	875	548	327	37.37%	-0.1238
3	(7.3500001430511475, 8.050000190734863]	773	509	264	34.15%	-0.0891
4	(8.050000190734863, 8.849999904632568]	1062	753	309	29.10%	-0.0697
5	(8.849999904632568, 9.25]	578	430	148	25.61%	-0.0218
6	(9.25, 10.050000190734863]	1235	982	253	20.49%	0.0138
7	(10.050000190734863, 11.449999809265137]	2187	1873	314	14.36%	0.1298
8	(11.449999809265137, 12.75]	1501	1362	139	9.26%	0.1674
9	(12.75, 13.550000190734863]	664	591	73	10.99%	0.0706
10	(13.550000190734863, inf]	1024	972	52	5.08%	0.1637

rare_field 稀疏字段(高缺失) — IV 0.0827，10 箱

分组号	分栏	Total	Good	Bad	Bad Rate	WOE
1	(-inf, -1.437000036239624]	23	21	2	8.70%	0.0847
2	(-1.437000036239624, -0.8540000021457672]	60	43	17	28.33%	-0.0735
3	(-0.8540000021457672, -0.443000003695488]	50	43	7	14.00%	0.1024
4	(-0.443000003695488, -0.25699999928474426]	25	15	10	40.00%	-0.1074
5	(-0.25699999928474426, -0.05299999937415123]	30	26	4	13.33%	0.0713
6	(-0.05299999937415123, 0.10999999940395355]	24	14	10	41.67%	-0.1134
7	(0.10999999940395355, 0.3699999898672104]	34	24	10	29.41%	-0.0554
8	(0.3699999898672104, 1.0099999904632568]	89	80	9	10.11%	0.2372
9	(1.0099999904632568, 1.324500024318695]	26	19	7	26.92%	-0.0279
10	(1.324500024318695, inf]	36	22	14	38.89%	-0.1369

avg_tnr 平均履约月数 — IV 0.0751，10 箱

分组号	分栏	Total	Good	Bad	Bad Rate	WOE
1	(-inf, 2.149999976158142]	926	474	452	48.81%	-0.2194
2	(2.149999976158142, 3.350000023841858]	961	609	352	36.63%	-0.1276
3	(3.350000023841858, 4.450000047683716]	1258	903	355	28.22%	-0.0706
4	(4.450000047683716, 4.8500001430511475]	550	422	128	23.27%	-0.0083
5	(4.8500001430511475, 5.450000047683716]	835	667	168	20.12%	0.0128
6	(5.450000047683716, 5.950000047683716]	748	580	168	22.46%	-0.0051
7	(5.950000047683716, 7.3500001430511475]	2097	1722	375	17.88%	0.0646
8	(7.3500001430511475, 8.550000190734863]	1478	1313	165	11.16%	0.1390
9	(8.550000190734863, 9.449999809265137]	754	657	97	12.86%	0.0651
10	(9.449999809265137, inf]	1194	1106	88	7.37%	0.1602

income_grade 收入等级 — IV 0.0632，5 箱

分组号	分栏	Total	Good	Bad	Bad Rate	WOE
1	(-inf, 3.5]	1211	743	468	38.65%	-0.1736
2	(3.5, 4.5]	2510	1766	744	29.64%	-0.1417
3	(4.5, 5.5]	3361	2689	672	19.99%	0.0398
4	(5.5, 6.5]	2503	2150	353	14.10%	0.1484
5	(6.5, inf]	1216	1105	111	9.13%	0.1419

repay_amt_6m_grade 近6月还款金额等级 — IV 0.0607，5 箱

分组号	分栏	Total	Good	Bad	Bad Rate	WOE
1	(-inf, 3.5]	1647	1047	600	36.43%	-0.1915
2	(3.5, 4.5]	2268	1634	634	27.95%	-0.1049
3	(4.5, 5.5]	2991	2384	607	20.29%	0.0305
4	(5.5, 6.5]	2286	1944	342	14.96%	0.1227
5	(6.5, inf]	1609	1444	165	10.25%	0.1624

query_cnt_3m 近3月机构查询次数 — IV 0.0483，9 箱

分组号	分栏	Total	Good	Bad	Bad Rate	WOE
1	(-inf, 0.5]	550	288	262	47.64%	-0.1355
2	(0.5, 1.5]	1123	724	399	35.53%	-0.1345
3	(1.5, 2.5]	1627	1151	476	29.26%	-0.0995
4	(2.5, 3.5]	1785	1384	401	22.46%	-0.0106
5	(3.5, 4.5]	1665	1365	300	18.02%	0.0523
6	(4.5, 5.5]	1347	1127	220	16.33%	0.0646
7	(5.5, 6.5]	1026	897	129	12.57%	0.0882
8	(6.5, 8.5]	1098	970	128	11.66%	0.1031
9	(8.5, inf]	580	547	33	5.69%	0.0940

bank_997_cnt_3m 近3月银行查询次数 — IV 0.0246，4 箱

分组号	分栏	Total	Good	Bad	Bad Rate	WOE
1	(-inf, 0.5]	3447	2899	548	15.90%	0.1392
2	(0.5, 1.5]	3739	2928	811	21.69%	0.0012
3	(1.5, 2.5]	2224	1655	569	25.58%	-0.0648
4	(2.5, inf]	1391	971	420	30.19%	-0.0990

card_usage_ratio 信用卡额度使用率 — IV 0.0183，10 箱

分组号	分栏	Total	Good	Bad	Bad Rate	WOE
1	(-inf, 0.107999999076128]	922	854	68	7.38%	0.1277
2	(0.107999999076128, 0.1578500047326088]	616	511	105	17.05%	0.0285
3	(0.1578500047326088, 0.3025500029325485]	2324	1848	476	20.48%	0.0224
4	(0.3025500029325485, 0.3381499946117401]	649	545	104	16.02%	0.0364
5	(0.3381499946117401, 0.4825499951839447]	2550	1996	554	21.73%	0.0003
6	(0.4825499951839447, 0.5283499956130981]	687	497	190	27.66%	-0.0388
7	(0.5283499956130981, 0.6137499809265137]	1203	941	262	21.78%	-0.0004
8	(0.6137499809265137, 0.6709499955177307]	594	405	189	31.82%	-0.0578
9	(0.6709499955177307, 0.7716499865055084]	715	520	195	27.27%	-0.0376
10	(0.7716499865055084, inf]	541	336	205	37.89%	-0.0844

tenure_m 授信时长(月) — IV 0.0170，10 箱

分组号	分栏	Total	Good	Bad	Bad Rate	WOE
1	(-inf, 2.5]	696	414	282	40.52%	-0.1218
2	(2.5, 10.5]	1179	891	288	24.43%	-0.0281
3	(10.5, 23.5]	2292	1813	479	20.90%	0.0148
4	(23.5, 30.5]	1310	997	313	23.89%	-0.0245
5	(30.5, 36.5]	1080	844	236	21.85%	-0.0011
6	(36.5, 39.5]	554	448	106	19.13%	0.0143
7	(39.5, 46.5]	1268	974	294	23.19%	-0.0161
8	(46.5, 53.5]	1211	970	241	19.90%	0.0199
9	(53.5, 58.5]	647	568	79	12.21%	0.0610
10	(58.5, inf]	564	534	30	5.32%	0.0937

edu_level 学历 — IV 0.0004，4 箱

分组号	分栏	Total	Good	Bad	Bad Rate	WOE
1	大专	3260	2574	686	21.04%	0.0155
2	本科	4550	3555	995	21.87%	-0.0035
3	硕士及以上	1405	1081	324	23.06%	-0.0160
4	高中及以下	1586	1243	343	21.63%	0.0015

region_code 区域 — IV 0.0004，5 箱

分组号	分栏	Total	Good	Bad	Bad Rate	WOE
1	其他	2119	1657	462	21.80%	-0.0011
2	华东	2156	1685	471	21.85%	-0.0018
3	华北	2119	1682	437	20.62%	0.0186
4	华南	2179	1701	478	21.94%	-0.0033
5	西南	2228	1728	500	22.44%	-0.0120

credit_no — IV 0.0000，2 箱

分组号	分栏	Total	Good	Bad	Bad Rate	WOE
1	C100000	1	1	0	0.00%	0.0002
2	Other	10800	8452	2348	21.74%	-0.0001

flag_special 特殊标记(高集中度) — IV 0.0000，1 箱

分组号	分栏	Total	Good	Bad	Bad Rate	WOE
1	(-inf, inf]	10801	8453	2348	21.74%	0.0000

data_flag — IV 0.0000，1 箱

分组号	分栏	Total	Good	Bad	Bad Rate	WOE
1	TRAIN	10801	8453	2348	21.74%	0.0000

注：统计全部候选变量的自动分箱明细（初分组），按 IV 降序展示前 18 个变量，共 18 个；完整数据见 feature_report.json。

04 单变量分析（粗分箱）

risk_score 内部风险评分 — 10 箱

分组号	分栏	Total	Good	Bad	Bad Rate	WOE
1	(-inf, 492.5]	1279	1232	47	3.67%	0.2165
2	(492.5, 519.5]	1276	1182	94	7.37%	0.1696
3	(519.5, 535.5]	1032	914	118	11.43%	0.1000
4	(535.5, 546.5]	801	694	107	13.36%	0.0648
5	(546.5, 564.5]	1350	1144	206	15.26%	0.0779
6	(564.5, 587.5]	1704	1320	384	22.54%	-0.0112
7	(587.5, 600.5]	870	626	244	28.05%	-0.0507
8	(600.5, 616.5]	797	531	266	33.38%	-0.0859
9	(616.5, 643.5]	955	529	426	44.61%	-0.1917
10	(643.5, inf]	737	281	456	61.87%	-0.2638

cust_seg2 客群分层等级 — 4 箱

分组号	分栏	Total	Good	Bad	Bad Rate	WOE
1	(-inf, 3.5]	2056	1867	189	9.19%	0.2166
2	(3.5, 4.5]	3333	2822	511	15.33%	0.1501
3	(4.5, 5.5]	3330	2505	825	24.77%	-0.0668
4	(5.5, inf]	2082	1259	823	39.53%	-0.2705

avg_tnr 平均履约月数 — 10 箱

分组号	分栏	Total	Good	Bad	Bad Rate	WOE
1	(-inf, 2.149999976158142]	926	474	452	48.81%	-0.2194
2	(2.149999976158142, 3.350000023841858]	961	609	352	36.63%	-0.1276
3	(3.350000023841858, 4.450000047683716]	1258	903	355	28.22%	-0.0706
4	(4.450000047683716, 4.8500001430511475]	550	422	128	23.27%	-0.0083
5	(4.8500001430511475, 5.450000047683716]	835	667	168	20.12%	0.0128
6	(5.450000047683716, 5.950000047683716]	748	580	168	22.46%	-0.0051
7	(5.950000047683716, 7.3500001430511475]	2097	1722	375	17.88%	0.0646
8	(7.3500001430511475, 8.550000190734863]	1478	1313	165	11.16%	0.1390
9	(8.550000190734863, 9.449999809265137]	754	657	97	12.86%	0.0651
10	(9.449999809265137, inf]	1194	1106	88	7.37%	0.1602

income_grade 收入等级 — 5 箱

分组号	分栏	Total	Good	Bad	Bad Rate	WOE
1	(-inf, 3.5]	1211	743	468	38.65%	-0.1736
2	(3.5, 4.5]	2510	1766	744	29.64%	-0.1417
3	(4.5, 5.5]	3361	2689	672	19.99%	0.0398
4	(5.5, 6.5]	2503	2150	353	14.10%	0.1484
5	(6.5, inf]	1216	1105	111	9.13%	0.1419

repay_amt_6m_grade 近6月还款金额等级 — 5 箱

分组号	分栏	Total	Good	Bad	Bad Rate	WOE
1	(-inf, 3.5]	1647	1047	600	36.43%	-0.1915
2	(3.5, 4.5]	2268	1634	634	27.95%	-0.1049
3	(4.5, 5.5]	2991	2384	607	20.29%	0.0305
4	(5.5, 6.5]	2286	1944	342	14.96%	0.1227
5	(6.5, inf]	1609	1444	165	10.25%	0.1624

query_cnt_3m 近3月机构查询次数 — 9 箱

分组号	分栏	Total	Good	Bad	Bad Rate	WOE
1	(-inf, 0.5]	550	288	262	47.64%	-0.1355
2	(0.5, 1.5]	1123	724	399	35.53%	-0.1345
3	(1.5, 2.5]	1627	1151	476	29.26%	-0.0995
4	(2.5, 3.5]	1785	1384	401	22.46%	-0.0106
5	(3.5, 4.5]	1665	1365	300	18.02%	0.0523
6	(4.5, 5.5]	1347	1127	220	16.33%	0.0646
7	(5.5, 6.5]	1026	897	129	12.57%	0.0882
8	(6.5, 8.5]	1098	970	128	11.66%	0.1031
9	(8.5, inf]	580	547	33	5.69%	0.0940

bank_997_cnt_3m 近3月银行查询次数 — 4 箱

分组号	分栏	Total	Good	Bad	Bad Rate	WOE
1	(-inf, 0.5]	3447	2899	548	15.90%	0.1392
2	(0.5, 1.5]	3739	2928	811	21.69%	0.0012
3	(1.5, 2.5]	2224	1655	569	25.58%	-0.0648
4	(2.5, inf]	1391	971	420	30.19%	-0.0990

注：统计入模变量经合并调整后的最终分箱明细（开发样本口径）。当前流程细/粗分箱由同一自动分箱算法直接产出。

05 变量筛选

项目	内容
筛选优先级	黑名单 > 白名单 > 自动规则
IV 下限	0.02
缺失率上限	0.8
集中度上限	0.9
相关性上限(皮尔逊)	0.7
PSI 上限	0.25
重要性 Top-K	80
黑名单	risk_score_v2
白名单	无

候选变量 19，成功分箱 18，入模 7

变量	白名单	入模	IV	缺失率	集中度	IV门槛	相关性	PSI	剔除原因
credit_no	—	N	0.0000	PASS	PASS	—	—	—	唯一率过高(ID列)
risk_score 内部风险评分	—	Y	0.1272	PASS	PASS	PASS	PASS	0.0037
cust_seg2 客群分层等级	—	Y	0.1060	PASS	PASS	PASS	PASS	0.0005
income_grade 收入等级	—	Y	0.0632	PASS	PASS	PASS	PASS	0.0009
bh_a001 百融履约指数A001	—	N	0.1128	PASS	PASS	PASS	FAIL	0.0027	与 risk_score 相关>0.7
bh_a009 百融信用历史长度	—	N	0.0991	PASS	PASS	PASS	FAIL	0.0012	与 risk_score 相关>0.7
bank_997_cnt_3m 近3月银行查询次数	—	Y	0.0246	PASS	PASS	PASS	PASS	0.0010
avg_tnr 平均履约月数	—	Y	0.0751	PASS	PASS	PASS	PASS	0.0030
repay_amt_6m_grade 近6月还款金额等级	—	Y	0.0607	PASS	PASS	PASS	PASS	0.0007
tenure_m 授信时长(月)	—	N	0.0170	PASS	PASS	FAIL	—	—	IV<0.02
query_cnt_3m 近3月机构查询次数	—	Y	0.0483	PASS	PASS	PASS	PASS	0.0014
card_usage_ratio 信用卡额度使用率	—	N	0.0183	PASS	PASS	FAIL	—	—	IV<0.02
rare_field 稀疏字段(高缺失)	—	N	0.0827	FAIL	PASS	—	—	—	缺失率>0.8
flag_special 特殊标记(高集中度)	—	N	0.0000	PASS	FAIL	—	—	—	集中度>0.9
edu_level 学历	—	N	0.0004	PASS	PASS	FAIL	—	—	IV<0.02
region_code 区域	—	N	0.0004	PASS	PASS	FAIL	—	—	IV<0.02
risk_score_v2 内部风险评分副本	—	N	—	—	—	—	—	—	黑名单
ScoreCust2_v2 客户评分卡2副本	—	N	0.1262	PASS	PASS	PASS	FAIL	0.0020	与 risk_score 相关>0.7
data_flag	—	N	0.0000	PASS	FAIL	—	—	—	集中度>0.9

06 回归结果

硬门槛：P≥0.1 / VIF>4.0 迭代剔除，共 1 个变量被剔除（2 轮）：cust_seg2。

Parameter	Label	Estimate(无惩罚)	Standard Error	Wald Chi-Square	Pr > ChiSq	Standardized Estimate
Intercept	Intercept	-1.3518	0.0268	2547.89	<0.0001	—
WOE_risk_score	risk_score	-5.5042	0.2873	367.17	<0.0001	-0.7633
WOE_avg_tnr	avg_tnr	-1.1249	0.2773	16.46	<0.0001	-0.1259
WOE_income_grade	income_grade	-0.5442	0.2263	5.78	0.0162	-0.0684
WOE_repay_amt_6m_grade	repay_amt_6m_grade	-0.4205	0.2304	3.33	0.0680	-0.0519
WOE_query_cnt_3m	query_cnt_3m	-0.6389	0.3394	3.54	0.0598	-0.0555
WOE_bank_997_cnt_3m	bank_997_cnt_3m	-0.7363	0.2968	6.15	0.0131	-0.0657

注：Estimate 列为无惩罚（精确推断）拟合；生产系数保持 L2 正则（model_results.json / score.py 口径），两套估计并列展示。标准误基于无惩罚似然观测信息矩阵。

07 VIF

VIF

Parameter	VIF
risk_score	2.68749
avg_tnr	1.72675
income_grade	1.33411
repay_amt_6m_grade	1.34148
query_cnt_3m	1.45984
bank_997_cnt_3m	1.06950

	risk_score	avg_tnr	income_grade	repay_amt_6m_grade	query_cnt_3m	bank_997_cnt_3m
risk_score	1.000	0.646	0.497	0.502	0.558	0.252
avg_tnr	0.646	1.000	0.343	0.350	0.392	0.178
income_grade	0.497	0.343	1.000	0.277	0.310	0.152
repay_amt_6m_grade	0.502	0.350	0.277	1.000	0.305	0.137
query_cnt_3m	0.558	0.392	0.310	0.305	1.000	0.155
bank_997_cnt_3m	0.252	0.178	0.152	0.137	0.155	1.000

08 最终评分卡

标准分600

标准ODD50

PDO20

变量	分栏	WOE	Coefficient	Scorecard Points
risk_score	(-inf, 492.5]	0.2165	-5.1699	120
	(492.5, 519.5]	0.1696	-5.1699	113
	(519.5, 535.5]	0.1000	-5.1699	103
	(535.5, 546.5]	0.0648	-5.1699	97
	(546.5, 564.5]	0.0779	-5.1699	99
	(564.5, 587.5]	-0.0112	-5.1699	86
	(587.5, 600.5]	-0.0507	-5.1699	80
	(600.5, 616.5]	-0.0859	-5.1699	75
	(616.5, 643.5]	-0.1917	-5.1699	59
	(643.5, inf]	-0.2638	-5.1699	48
avg_tnr	(-inf, 2.149999976158142]	-0.2194	-1.2598	80
	(2.149999976158142, 3.350000023841858]	-0.1276	-1.2598	83
	(3.350000023841858, 4.450000047683716]	-0.0706	-1.2598	85
	(4.450000047683716, 4.8500001430511475]	-0.0083	-1.2598	87
	(4.8500001430511475, 5.450000047683716]	0.0128	-1.2598	88
	(5.450000047683716, 5.950000047683716]	-0.0051	-1.2598	87
	(5.950000047683716, 7.3500001430511475]	0.0646	-1.2598	90
	(7.3500001430511475, 8.550000190734863]	0.1390	-1.2598	93
	(8.550000190734863, 9.449999809265137]	0.0651	-1.2598	90
	(9.449999809265137, inf]	0.1602	-1.2598	93
income_grade	(-inf, 3.5]	-0.1736	-0.6188	85
	(3.5, 4.5]	-0.1417	-0.6188	85
	(4.5, 5.5]	0.0398	-0.6188	88
	(5.5, 6.5]	0.1484	-0.6188	90
	(6.5, inf]	0.1419	-0.6188	90
repay_amt_6m_grade	(-inf, 3.5]	-0.1915	-0.5006	85
	(3.5, 4.5]	-0.1049	-0.5006	86
	(4.5, 5.5]	0.0305	-0.5006	88
	(5.5, 6.5]	0.1227	-0.5006	89
	(6.5, inf]	0.1624	-0.5006	90
query_cnt_3m	(-inf, 0.5]	-0.1355	-0.6938	85
	(0.5, 1.5]	-0.1345	-0.6938	85
	(1.5, 2.5]	-0.0995	-0.6938	86
	(2.5, 3.5]	-0.0106	-0.6938	87
	(3.5, 4.5]	0.0523	-0.6938	89
	(4.5, 5.5]	0.0646	-0.6938	89
	(5.5, 6.5]	0.0882	-0.6938	89
	(6.5, 8.5]	0.1031	-0.6938	90
	(8.5, inf]	0.0940	-0.6938	90
bank_997_cnt_3m	(-inf, 0.5]	0.1392	-0.7190	91
	(0.5, 1.5]	0.0012	-0.7190	88
	(1.5, 2.5]	-0.0648	-0.7190	86
	(2.5, inf]	-0.0990	-0.7190	86

09 模型表现

模型区分度评估（按样本）

模型	样本	KS	AUC	GINI
LR + WOE 评分卡	开发样本	0.3748	0.7541	0.5082
LR + WOE 评分卡	测试样本	0.3641	0.7455	0.4910
LR + WOE 评分卡	训练+测试样本	0.3696	0.7507	0.5014
LR + WOE 评分卡	跨时间验证样本	0.3433	0.7312	0.4624
LR + WOE 评分卡	PSI时点样本	0.3739	0.7468	0.4936

LR + WOE 评分卡 — 4 样本评估表（score_cut 粒度）

开发样本（KS 0.3748，AUC 0.7541）

cuts	cnt	bad	good	bad_rate	population_pct	cum_bad_rate	cum_good_rate
(0.0469, 0.0619]	1081	43	1038	3.98%	10.01%	1.83%	12.28%
(0.0619, 0.0821]	1080	68	1012	6.30%	10.00%	4.73%	24.25%
(0.0822, 0.1124]	1080	97	983	8.98%	10.00%	8.86%	35.88%
(0.1125, 0.1333]	1080	138	942	12.78%	10.00%	14.74%	47.02%
(0.1334, 0.1593]	1080	167	913	15.46%	10.00%	21.85%	57.83%
(0.1593, 0.2090]	1083	211	872	19.48%	10.03%	30.83%	68.14%
(0.2090, 0.2554]	1077	271	806	25.16%	9.97%	42.38%	77.68%
(0.2554, 0.3219]	1080	300	780	27.78%	10.00%	55.15%	86.90%
(0.3221, 0.4915]	1080	427	653	39.54%	10.00%	73.34%	94.63%
(0.4916, 0.6600]	1080	626	454	57.96%	10.00%	100.00%	100.00%

测试样本（KS 0.3641，AUC 0.7455）

cuts	cnt	bad	good	bad_rate	population_pct	cum_bad_rate	cum_good_rate
(0.0469, 0.0611]	720	30	690	4.17%	10.00%	1.92%	12.24%
(0.0611, 0.0811]	721	48	673	6.66%	10.02%	4.99%	24.19%
(0.0812, 0.1129]	719	71	648	9.87%	9.99%	9.53%	35.69%
(0.1130, 0.1318]	720	95	625	13.19%	10.00%	15.60%	46.78%
(0.1319, 0.1561]	720	104	616	14.44%	10.00%	22.25%	57.71%
(0.1561, 0.2076]	720	152	568	21.11%	10.00%	31.97%	67.79%
(0.2076, 0.2561]	719	165	554	22.95%	9.99%	42.52%	77.62%
(0.2561, 0.3198]	720	222	498	30.83%	10.00%	56.71%	86.46%
(0.3199, 0.4821]	720	273	447	37.92%	10.00%	74.17%	94.39%
(0.4822, 0.6600]	720	404	316	56.11%	10.00%	100.00%	100.00%

训练+测试样本（KS 0.3696，AUC 0.7507）

cuts	cnt	bad	good	bad_rate	population_pct	cum_bad_rate	cum_good_rate
(0.0469, 0.0616]	1800	73	1727	4.06%	10.00%	1.87%	12.26%
(0.0616, 0.0818]	1800	114	1686	6.33%	10.00%	4.78%	24.23%
(0.0818, 0.1128]	1802	169	1633	9.38%	10.01%	9.10%	35.82%
(0.1128, 0.1327]	1798	234	1564	13.01%	9.99%	15.08%	46.92%
(0.1327, 0.1579]	1800	279	1521	15.50%	10.00%	22.21%	57.72%
(0.1579, 0.2085]	1802	354	1448	19.64%	10.01%	31.26%	67.99%
(0.2086, 0.2556]	1798	437	1361	24.30%	9.99%	42.43%	77.65%
(0.2557, 0.3207]	1800	521	1279	28.94%	10.00%	55.75%	86.73%
(0.3208, 0.4869]	1800	703	1097	39.06%	10.00%	73.72%	94.52%
(0.4869, 0.6600]	1800	1028	772	57.11%	10.00%	100.00%	100.00%

跨时间验证样本（KS 0.3433，AUC 0.7312）

cuts	cnt	bad	good	bad_rate	population_pct	cum_bad_rate	cum_good_rate
(0.0469, 0.0624]	600	23	577	3.83%	10.00%	1.67%	12.48%
(0.0624, 0.0821]	600	54	546	9.00%	10.00%	5.60%	24.29%
(0.0821, 0.1121]	600	63	537	10.50%	10.00%	10.17%	35.90%
(0.1122, 0.1335]	600	90	510	15.00%	10.00%	16.72%	46.93%
(0.1335, 0.1615]	601	110	491	18.30%	10.02%	24.71%	57.55%
(0.1615, 0.2088]	599	124	475	20.70%	9.98%	33.72%	67.82%
(0.2089, 0.2536]	600	160	440	26.67%	10.00%	45.35%	77.34%
(0.2536, 0.3221]	600	197	403	32.83%	10.00%	59.67%	86.05%
(0.3225, 0.4864]	600	238	362	39.67%	10.00%	76.96%	93.88%
(0.4865, 0.6600]	600	317	283	52.83%	10.00%	100.00%	100.00%

注：score_cut 表基于模型输出 PD 等频分 10 档（分数升序），每表附 KS/AUC；训练+测试 = TRAIN 全体。

10 KS和GINI

LR + WOE 评分卡

样本	KS	GINI	AUC
开发样本	0.3748	0.5082	0.7541
测试样本	0.3641	0.4910	0.7455
训练+测试样本	0.3696	0.5014	0.7507
跨时间验证样本	0.3433	0.4624	0.7312
PSI时点样本	0.3739	0.4936	0.7468

开发样本

测试样本

训练+测试样本

跨时间验证样本

注：GINI = 2×AUC−1。KS 图 X 轴为分数（升序），红虚线为最大间隔位置；ROC 图例标注 AUC。

11 PSI

模型稳定性评估

模型	训练集与测试集	训练集与跨时间验证	训练集与oot验证集	训练集与psi验证集
LR + WOE 评分卡	0.0032	0.0011	—	0.0032

分数段 PSI 明细（开发样本 vs 跨时间验证样本）

分数段	开发样本占比	跨时间验证样本占比	PSI 贡献	累计 PSI
(-inf,0.0619]	9.99%	9.68%	0.0001	0.0001
(0.0619,0.0821]	10.01%	10.33%	0.0001	0.0002
(0.0821,0.1124]	10.01%	10.12%	0.0000	0.0002
(0.1124,0.1333]	9.99%	9.75%	0.0001	0.0003
(0.1333,0.1593]	10.00%	9.77%	0.0001	0.0003
(0.1593,0.209]	9.99%	10.37%	0.0001	0.0005
(0.209,0.2554]	10.01%	10.30%	0.0001	0.0005
(0.2554,0.3219]	10.00%	9.67%	0.0001	0.0007
(0.3219,0.4915]	9.99%	10.47%	0.0002	0.0009
(0.4915,inf]	10.02%	9.55%	0.0002	0.0011

分数段 PSI 明细（开发样本 vs PSI时点样本）

分数段	开发样本占比	PSI时点占比	PSI 贡献	累计 PSI
(-inf,0.0619]	9.99%	9.60%	0.0002	0.0002
(0.0619,0.0821]	10.01%	10.00%	0.0000	0.0002
(0.0821,0.1124]	10.01%	10.55%	0.0003	0.0004
(0.1124,0.1333]	9.99%	10.15%	0.0000	0.0005
(0.1333,0.1593]	10.00%	10.08%	0.0000	0.0005
(0.1593,0.209]	9.99%	10.90%	0.0008	0.0013
(0.209,0.2554]	10.01%	8.90%	0.0013	0.0026
(0.2554,0.3219]	10.00%	9.42%	0.0003	0.0029
(0.3219,0.4915]	9.99%	10.52%	0.0003	0.0032
(0.4915,inf]	10.02%	9.88%	0.0000	0.0032

模型样本量分布评估

评分段	开发样本样本量	开发样本占比	测试样本样本量	测试样本占比	训练+测试样本样本量	训练+测试样本占比	跨时间验证样本样本量	跨时间验证样本占比	PSI时点样本样本量	PSI时点样本占比
(-inf, inf]	1079	9.99%	757	10.52%	1836	10.20%	581	9.68%	576	9.60%
(0.0619, 0.0821]	1081	10.01%	712	9.89%	1793	9.96%	620	10.33%	600	10.00%
(0.0821, 0.1124]	1081	10.01%	673	9.35%	1754	9.74%	607	10.12%	633	10.55%
(0.1124, 0.1333]	1079	9.99%	784	10.89%	1863	10.35%	585	9.75%	609	10.15%
(0.1333, 0.1593]	1080	10.00%	735	10.21%	1815	10.08%	586	9.77%	605	10.08%
(0.1593, 0.209]	1079	9.99%	678	9.42%	1757	9.76%	622	10.37%	654	10.90%
(0.209, 0.2554]	1081	10.01%	680	9.45%	1761	9.78%	618	10.30%	534	8.90%
(0.2554, 0.3219]	1080	10.00%	762	10.58%	1842	10.23%	580	9.67%	565	9.42%
(0.3219, 0.4915]	1079	9.99%	747	10.38%	1826	10.14%	628	10.47%	631	10.52%
(0.4915, inf]	1082	10.02%	671	9.32%	1753	9.74%	573	9.55%	593	9.88%
sum	10801	100.00%	7199	100.00%	18000	100.00%	6000	100.00%	6000	100.00%

变量 PSI（开发样本 vs PSI时点样本，按入模变量分箱）

特征项	PSI	状态
risk_score 内部风险评分	0.0037	稳定
avg_tnr 平均履约月数	0.0030	稳定
query_cnt_3m 近3月机构查询次数	0.0014	稳定
bank_997_cnt_3m 近3月银行查询次数	0.0010	稳定
income_grade 收入等级	0.0009	稳定
repay_amt_6m_grade 近6月还款金额等级	0.0007	稳定
cust_seg2 客群分层等级	0.0005	稳定

risk_score — PSI 0.0037

分栏	开发基准比例	PSI时点样本比例	变更	贡献
(-inf, 492.5]	11.84%	10.97%	-0.87%	0.0007
(492.5, 519.5]	11.81%	12.98%	1.17%	0.0011
(519.5, 535.5]	9.55%	9.55%	-0.00%	0.0000
(535.5, 546.5]	7.42%	8.02%	0.60%	0.0005
(546.5, 564.5]	12.50%	12.57%	0.07%	0.0000
(564.5, 587.5]	15.78%	14.98%	-0.79%	0.0004
(587.5, 600.5]	8.05%	7.53%	-0.52%	0.0003
(600.5, 616.5]	7.38%	6.93%	-0.45%	0.0003
(616.5, 643.5]	8.84%	9.43%	0.59%	0.0004
(643.5, inf]	6.82%	7.03%	0.21%	0.0001

cust_seg2 — PSI 0.0005

分栏	开发基准比例	PSI时点样本比例	变更	贡献
(-inf, 3.5]	19.04%	19.07%	0.03%	0.0000
(3.5, 4.5]	30.86%	30.50%	-0.36%	0.0000
(4.5, 5.5]	30.83%	31.77%	0.94%	0.0003
(5.5, inf]	19.28%	18.67%	-0.61%	0.0002

avg_tnr — PSI 0.0030

分栏	开发基准比例	PSI时点样本比例	变更	贡献
(-inf, 2.149999976158142]	8.57%	8.88%	0.31%	0.0001
(2.149999976158142, 3.350000023841858]	8.90%	8.87%	-0.03%	0.0000
(3.350000023841858, 4.450000047683716]	11.65%	11.57%	-0.08%	0.0000
(4.450000047683716, 4.8500001430511475]	5.09%	5.18%	0.09%	0.0000
(4.8500001430511475, 5.450000047683716]	7.73%	7.87%	0.14%	0.0000
(5.450000047683716, 5.950000047683716]	6.93%	6.62%	-0.31%	0.0001
(5.950000047683716, 7.3500001430511475]	19.41%	18.15%	-1.26%	0.0009
(7.3500001430511475, 8.550000190734863]	13.68%	13.52%	-0.17%	0.0000
(8.550000190734863, 9.449999809265137]	6.98%	8.15%	1.17%	0.0018
(9.449999809265137, inf]	11.05%	11.20%	0.15%	0.0000

income_grade — PSI 0.0009

分栏	开发基准比例	PSI时点样本比例	变更	贡献
(-inf, 3.5]	11.21%	10.70%	-0.51%	0.0002
(3.5, 4.5]	23.24%	23.15%	-0.09%	0.0000
(4.5, 5.5]	31.12%	31.07%	-0.05%	0.0000
(5.5, 6.5]	23.17%	24.25%	1.08%	0.0005
(6.5, inf]	11.26%	10.83%	-0.42%	0.0002

repay_amt_6m_grade — PSI 0.0007

分栏	开发基准比例	PSI时点样本比例	变更	贡献
(-inf, 3.5]	15.25%	14.83%	-0.42%	0.0001
(3.5, 4.5]	21.00%	21.75%	0.75%	0.0003
(4.5, 5.5]	27.69%	26.92%	-0.78%	0.0002
(5.5, 6.5]	21.16%	21.43%	0.27%	0.0000
(6.5, inf]	14.90%	15.07%	0.17%	0.0000

query_cnt_3m — PSI 0.0014

分栏	开发基准比例	PSI时点样本比例	变更	贡献
(-inf, 0.5]	5.09%	5.28%	0.19%	0.0001
(0.5, 1.5]	10.40%	10.45%	0.05%	0.0000
(1.5, 2.5]	15.06%	14.95%	-0.11%	0.0000
(2.5, 3.5]	16.53%	16.27%	-0.26%	0.0000
(3.5, 4.5]	15.42%	15.28%	-0.13%	0.0000
(4.5, 5.5]	12.47%	13.08%	0.61%	0.0003
(5.5, 6.5]	9.50%	9.20%	-0.30%	0.0001
(6.5, 8.5]	10.17%	10.67%	0.50%	0.0002
(8.5, inf]	5.37%	4.82%	-0.55%	0.0006

bank_997_cnt_3m — PSI 0.0010

分栏	开发基准比例	PSI时点样本比例	变更	贡献
(-inf, 0.5]	31.91%	32.48%	0.57%	0.0001
(0.5, 1.5]	34.62%	35.27%	0.65%	0.0001
(1.5, 2.5]	20.59%	20.35%	-0.24%	0.0000
(2.5, inf]	12.88%	11.90%	-0.98%	0.0008

注：PSI ≤ 0.1 视为稳定，0.1~0.25 需关注，> 0.25 不稳定。

12 单变量分析

注：统计所有入模变量的单变量分析（开发样本口径）。预留验证/总体/OOT 口径的分组明细待扩展。

risk_score 内部风险评分

组号	Value	Tot. Records	累计占比	#Goods	#Bads	Bad Rate	G/B Odds
1	(-inf, 492.5]	1279	11.84%	1232	47	3.67%	26.21
2	(492.5, 519.5]	1276	23.66%	1182	94	7.37%	12.57
3	(519.5, 535.5]	1032	33.21%	914	118	11.43%	7.75
4	(535.5, 546.5]	801	40.63%	694	107	13.36%	6.49
5	(546.5, 564.5]	1350	53.12%	1144	206	15.26%	5.55
6	(564.5, 587.5]	1704	68.90%	1320	384	22.54%	3.44
7	(587.5, 600.5]	870	76.96%	626	244	28.05%	2.57
8	(600.5, 616.5]	797	84.33%	531	266	33.38%	2.00
9	(616.5, 643.5]	955	93.18%	529	426	44.61%	1.24
10	(643.5, inf]	737	100.00%	281	456	61.87%	0.62
合计		10801		8453	2348	21.74%	3.60

cust_seg2 客群分层等级

组号	Value	Tot. Records	累计占比	#Goods	#Bads	Bad Rate	G/B Odds
1	(-inf, 3.5]	2056	19.04%	1867	189	9.19%	9.88
2	(3.5, 4.5]	3333	49.89%	2822	511	15.33%	5.52
3	(4.5, 5.5]	3330	80.72%	2505	825	24.77%	3.04
4	(5.5, inf]	2082	100.00%	1259	823	39.53%	1.53
合计		10801		8453	2348	21.74%	3.60

avg_tnr 平均履约月数

组号	Value	Tot. Records	累计占比	#Goods	#Bads	Bad Rate	G/B Odds
1	(-inf, 2.149999976158142]	926	8.57%	474	452	48.81%	1.05
2	(2.149999976158142, 3.350000023841858]	961	17.47%	609	352	36.63%	1.73
3	(3.350000023841858, 4.450000047683716]	1258	29.12%	903	355	28.22%	2.54
4	(4.450000047683716, 4.8500001430511475]	550	34.21%	422	128	23.27%	3.30
5	(4.8500001430511475, 5.450000047683716]	835	41.94%	667	168	20.12%	3.97
6	(5.450000047683716, 5.950000047683716]	748	48.87%	580	168	22.46%	3.45
7	(5.950000047683716, 7.3500001430511475]	2097	68.28%	1722	375	17.88%	4.59
8	(7.3500001430511475, 8.550000190734863]	1478	81.96%	1313	165	11.16%	7.96
9	(8.550000190734863, 9.449999809265137]	754	88.95%	657	97	12.86%	6.77
10	(9.449999809265137, inf]	1194	100.00%	1106	88	7.37%	12.57
合计		10801		8453	2348	21.74%	3.60

income_grade 收入等级

组号	Value	Tot. Records	累计占比	#Goods	#Bads	Bad Rate	G/B Odds
1	(-inf, 3.5]	1211	11.21%	743	468	38.65%	1.59
2	(3.5, 4.5]	2510	34.45%	1766	744	29.64%	2.37
3	(4.5, 5.5]	3361	65.57%	2689	672	19.99%	4.00
4	(5.5, 6.5]	2503	88.74%	2150	353	14.10%	6.09
5	(6.5, inf]	1216	100.00%	1105	111	9.13%	9.95
合计		10801		8453	2348	21.74%	3.60

repay_amt_6m_grade 近6月还款金额等级

组号	Value	Tot. Records	累计占比	#Goods	#Bads	Bad Rate	G/B Odds
1	(-inf, 3.5]	1647	15.25%	1047	600	36.43%	1.75
2	(3.5, 4.5]	2268	36.25%	1634	634	27.95%	2.58
3	(4.5, 5.5]	2991	63.94%	2384	607	20.29%	3.93
4	(5.5, 6.5]	2286	85.10%	1944	342	14.96%	5.68
5	(6.5, inf]	1609	100.00%	1444	165	10.25%	8.75
合计		10801		8453	2348	21.74%	3.60

query_cnt_3m 近3月机构查询次数

组号	Value	Tot. Records	累计占比	#Goods	#Bads	Bad Rate	G/B Odds
1	(-inf, 0.5]	550	5.09%	288	262	47.64%	1.10
2	(0.5, 1.5]	1123	15.49%	724	399	35.53%	1.81
3	(1.5, 2.5]	1627	30.55%	1151	476	29.26%	2.42
4	(2.5, 3.5]	1785	47.08%	1384	401	22.46%	3.45
5	(3.5, 4.5]	1665	62.49%	1365	300	18.02%	4.55
6	(4.5, 5.5]	1347	74.97%	1127	220	16.33%	5.12
7	(5.5, 6.5]	1026	84.46%	897	129	12.57%	6.95
8	(6.5, 8.5]	1098	94.63%	970	128	11.66%	7.58
9	(8.5, inf]	580	100.00%	547	33	5.69%	16.58
合计		10801		8453	2348	21.74%	3.60

bank_997_cnt_3m 近3月银行查询次数

组号	Value	Tot. Records	累计占比	#Goods	#Bads	Bad Rate	G/B Odds
1	(-inf, 0.5]	3447	31.91%	2899	548	15.90%	5.29
2	(0.5, 1.5]	3739	66.53%	2928	811	21.69%	3.61
3	(1.5, 2.5]	2224	87.12%	1655	569	25.58%	2.91
4	(2.5, inf]	1391	100.00%	971	420	30.19%	2.31
合计		10801		8453	2348	21.74%	3.60

13 压力测试

注：验证压力情境下模型的有效性和区分能力稳定性：a.坏浓度更高样本 b.时间外样本 c.特定特征客群 d.极端时间样本。当前流程自动生成情景1（跨时间样本）与情景2（跨时间样本剔除一半好样本，批注16）。

测试情景1：跨时间验证样本

变量贡献度

变量	贡献度
risk_score	0.7633
avg_tnr	0.1259
income_grade	0.0684
bank_997_cnt_3m	0.0657
query_cnt_3m	0.0555
repay_amt_6m_grade	0.0519

区分度验证 — LR + WOE 评分卡 @ 跨时间样本

KS = 0.3433，GINI = 0.4624

组号	分数区间(PD)	样本数	累计占比	Good	Bad	累计Good占比	累计Bad占比	分组坏率	累计坏率	分组Odds	累计Odds
1	0.0469 ~ 0.0624	600	10.00%	577	23	12.48%	1.67%	3.83%	3.83%	25.09	25.09
2	0.0624 ~ 0.0821	600	20.00%	546	54	11.81%	5.60%	9.00%	6.42%	10.11	14.58
3	0.0821 ~ 0.1121	600	30.00%	537	63	11.61%	10.17%	10.50%	7.78%	8.52	11.86
4	0.1122 ~ 0.1335	600	40.00%	510	90	11.03%	16.72%	15.00%	9.58%	5.67	9.43
5	0.1335 ~ 0.1615	599	49.98%	489	110	10.58%	24.71%	18.36%	11.34%	4.45	7.82
6	0.1615 ~ 0.2088	601	60.00%	477	124	10.32%	33.72%	20.63%	12.89%	3.85	6.76
7	0.2089 ~ 0.2536	600	70.00%	440	160	9.52%	45.35%	26.67%	14.86%	2.75	5.73
8	0.2536 ~ 0.3221	600	80.00%	403	197	8.72%	59.67%	32.83%	17.10%	2.05	4.85
9	0.3225 ~ 0.4864	600	90.00%	362	238	7.83%	76.96%	39.67%	19.61%	1.52	4.10
10	0.4865 ~ 0.6600	600	100.00%	283	317	6.12%	100.00%	52.83%	22.93%	0.89	3.36
TOTALS		6000		4624	1376			22.93%		—	3.36

测试情景2：跨时间样本剔除一半好样本

指标	数值
样本量	6000 → 3688（剔除好样本 2312）
KS	0.3377
AUC	0.7280
GINI	0.4560

注：情景2 模拟好样本分布右移的压力情形（固定 seed=42，可复现）；剔除后 KS/AUC 下降幅度反映模型对好样本结构的敏感度。