Design & media

China Insurance Actuarial Pricing Expert

Try it

China insurance actuarial pricing reference using the 4th Life Table (2025) and C-ROSS Phase II, with formulas and Python templates.

What it does

An educational reference and analytical framework for China insurance actuarial pricing. Built around the 4th Life Table (2025, CL1/CL2/CL3, mandatory January 1, 2026) and C-ROSS Phase II (Rules II 2024). Provides formulas and Python code templates for pure premium, reserves (UEPR / IBNR / long-term liability), solvency capital, critical illness pricing, group insurance experience rating, and IFRS 17 / HKFRS 17 fulfillment cash flow modeling. Ships six reference files on life table usage, pricing formulas, critical illness, reserves, and group insurance. Reference material only; not financial, legal, or insurance advice; outputs require human review.

When to use it

  • Pricing a critical illness product with CL1/CL2/CL3
  • Calculating reserves under C-ROSS Phase II
  • Modeling IFRS 17 fulfillment cash flows
  • Rating group insurance by experience

The skill document

China Insurance Actuarial Pricing Expert (4th Life Table 2025) / �й����㶨��ʦ��������������2025��|

⚠️ SECURITY NOTICE

  • Type: Educational reference / analytical framework ONLY
  • No executable code, scripts, or binaries included
  • No persistent storage, network calls, or background execution
  • No credential collection, PII processing, or system access
  • All outputs require human review before real-world application
  • NOT financial, legal, or insurance advice

保险监管最新动态 [2026-06-15更新]

动态类型内容摘要发布时间影响范围
监管发布NFRA 2026年第2号令:《银行保险机构许可证管理办法》6月1日施行,取消保险许可证2026-06精算定价模型需更新费用假设(新增许可证管理岗位成本)
监管发布许可证换证过渡期2026.6-2028.5,险企须设专职许可证管理岗位2026-06产品备案流程更新
监管动态2026年Q1监管处罚分析:分级分类处罚标准+个人追责条款落地2026-Q1合规风险管理

数据截止: 2026-06-15 | 来源:国家金融监督管理总局、政府网、金融新闻网 声明: 以上动态供参考,具体以官方最新发布为准

English: AI-powered China insurance actuarial pricing expert �� the definitive skill for Chinese actuaries and product pricing teams. Built on the 4th Life Table (2025, effective 2026-01-01) and C-ROSS Phase II framework. Covers pure premium calculation, reserve calculation, solvency capital assessment, and insurance product pricing. Delivers production-ready Python pricing code.

����: �й����㶨��ʦ�������ڵ�������������2025�귢����2026��1��1��ʵʩ���ͳ��������ڹ��̣�C-ROSS Rules II���Ĵ�ֱ����Skill�����Ǵ����Ѽ��㡢׼������㡢�������ʱ�ռ���붨��Эͬ�����ն���ר������������㶨��Python���롣���ã�����ʦ����Ʒ���۸ڡ����ղ�Ʒ���������վ���ʵϰ����������ѯ��


Trigger Keywords / �����ؼ���

English Triggers: actuarial, pricing, insurance pricing, life table, mortality table, critical illness rate, experience rate, reserve calculation, solvency capital, C-ROSS, China actuarial, product design, insurance product, 4th life table

���Ĵ����ʣ����ȣ��� ���㡢���㶨�ۡ������嶨���������������ʱ��������ʱ����ؼ������ʡ����鷢���ʡ�׼������㡢���������ʱ������ղ�Ʒ���ۡ�CASS���й�����ʦЭ�ᡢ���������������ؼ������ӡ���Ʒ���


保险监管最新动态 [2026-06-28更新]

动态类型内容摘要发布时间影响范围
监管发布金融监管总局发布《关于银行业保险业人工智能安全开发应用的指导意见》,界定承保理赔、风险管理等为AI高风险应用场景2026-06-18保险AI应用合规与风险管控
监管计划NFRA发布《2026年规章制定工作计划》:保险资金运用管理办法、偿付能力管理办法纳入修订,流动性风险、网络安全管理办法新制定2026-06-23保险监管全链条
监管施行NFRA 2026年第2号令《银行保险机构许可证管理办法》6月1日起施行,取消保险许可证统一为金融许可证2026-06-01保险业务资质与合规管理

数据截止: 2026-06-28 | 来源:国家金融监督管理总局、行业公开信息 声明: 以上动态供参考,具体以官方最新发布为准

Core Capabilities / ��������|

0. 2025-2026 Latest Regulatory Updates / ���¼�ܶ�̬������2026��5�£�|

ʱ����������Ӱ��
2025��10��29��������������������CL1/CL2/CL3��2026��1��1����ǿ��ʹ�ã�������/�����ֵȫ������
2024��3��18�������������ȫ��ʵʩ����ʱ������ʷ��ա��ؼ�������ȫ������
2024����Ԥ�����������µ���3.0%��ͳ�ն��۳ɱ��½����ֺ��վ������������
2025����IFRS 17/HKFRS 17��A�����������ƹ�׼��������߼������CSM̯��Ҫ������
2026��1��1��������������ǿ��ִ�����������ղ�Ʒ�밴�±����¶��ۻ�����

1. 4th Life Table (2025) Full Analysis / ��������������2025��ȫ����|

����������

  • 2025��10��29�գ��й�����ʦЭ����ʽ����
  • ���ڼ���ܾ�ͬ������ʵʩ֪ͨ
  • ��2026��1��1����ʵʩ�����������ڣ�
  • ��������������������ף�2010��汾������ӳ�й��˾������ӳ������仯����|

���ź��ı�|

����Ӣ����;
��������ҵ��һ��CL1�������ա��������յȲ�Ʒ����
��������ҵ�����CL2��ȫ�ա�����յȲ�Ʒ����
������ҵ���CL3ר�����ϱ��ա��������Ȳ�Ʒ����

���������������Ҫ�仯|

ά�������ף�2010�棩�����ף�2025�棩Ӱ��
�����ڼ�2005-2008��2018-2023������ӳ��ǰ����ˮƽ
����Ԥ��������60�꣩+20.7��+22.5��������⸶���ӳ�
Ů��Ԥ��������60�꣩+24.2��+26.1��Ů�������ճɱ�����
�����������ࣩ��ר������ר��������ն��۸���׼

2. China Actuarial Pricing Methodology / �й����㶨�۷�����|

���ղ�Ʒ���ۻ�����ʽ|

ë���� = ������ �� (1 + ���ӷ�����)
������ = ���ɴ����� / �����ֵϵ��
      = ���ս�� �� ƽ�������� �� ƽ�������ڼ� �� ����ϵ��

���ղ�Ʒ����Ҫ��|

Ҫ��˵������ȡֵ
Ԥ����������Դ��������������CL1/CL2/CL3
Ԥ���������չ�˾Ͷ�������ʼ���2.5%-3.5%��2024����޸�3%��
Ԥ����������������/����/�������10%-35%�����������
Ԥ����������˾����Ҫ��5%-15%
���ѽ��ɷ�ʽ����/�ڽ�/�꽻�ڽ�������

�ؼ��ն��۷�����Pythonʾ����|

# �ؼ��մ����Ѽ�����
def critical_illness_premium(age, sum_assured, policy_term, payment_term):
    # ����������������������ҵ�����(CL2)Ϊ����
    i = 0.025  # Ԥ�����ʣ�3.0%���޺�ı��ؼ��裩
    v = 1 / (1 + i)
    
    # �ؼ�������ֵ
    A_crit = 0
    for t in range(policy_term):
        q_crit_t = lookup_critical_illness_rate(age + t, t)
        A_crit += v**(t+1) * q_crit_t * sum_assured
    
    # ����������ֵ������/ȫ�У�
    A_death = 0
    for t in range(policy_term):
        q_death_t = lookup_mortality_rate_CL2(age + t, t)
        A_death += v**(t+1) * q_death_t * sum_assured
    
    # ���������ֵ���ɷ��ڣ�
    ?x_n = sum(v**t * survival_rate(age, t) for t in range(payment_term))
    
    # ������ = (�ؼ�������ֵ + ����������ֵ) / ���������ֵ
    pure_premium = (A_crit + A_death) / ?x_n
    return pure_premium

3. Reserve Calculation / ׼�������|

���������������
δ��������׼����UEPR��Ϊδ���⸶׼����Ǯʣ�ౣ�����ڵĴ�������ֵ
�ѷ������׼����IBNR���ѷ���δ�������������ƣ����������εȷ�����
����׼�������ѳ����Բ��Ժ󲹳��ֽ�������
��������׼�������ٱ��ճ��ڸ�ծ������������ƽ������

4. Group Insurance Pricing / ���ն���ר��|

���ն���Ҫ��|

Ҫ��˵����������
�����ģ�α���������Խ�࣬����ԽС
��ҵ������ҵ������ҵ��ҵ���յȼ�ϵ��
����ṹԱ��ƽ������/�ֲ�Ա��ƽ������Խ�󣬱���Խ��
��ʷ�⸶��ȥ1-3���⸶��¼������ʵ���
������������Ϸ�Χ/����/�����������Ӱ��

���վ�����ʼ���|

def group_experience_rate(base_premium, experience_factor):
    if experience_factor < 0.7:
        rate_adjustment = 0.85  # �������飬����15%
    elif experience_factor < 0.9:
        rate_adjustment = 0.95  # �Ϻþ��飬����5%
    elif experience_factor < 1.1:
        rate_adjustment = 1.00  # ��׼����
    elif experience_factor < 1.3:
        rate_adjustment = 1.15  # �ϲ�飬�ӷ�15%
    else:
        rate_adjustment = 1.30  # ���Ӿ��飬�ӷ�30%
    return base_premium * rate_adjustment

5. IFRS 17 / HKFRS 17 �����¹棨A�������������ã�|

IFRS 17 �Ծ��㶨�۵ĺ���Ӱ��

��ĿԭIFRS 4IFRS 17 �仯����Ӧ��
��������ȷ���ڼ䱣�����շ������루�DZ��ѣ����ѷֽ�Ϊ�⸶/����/CSM�ͷ�
׼���������ʷ�ɱ������г�һ�²�����Ҫ�������������
��ͬ����߼�CSM������������δ������ÿ���ͷţ�����һ��ȷ��
���յ���RA��ר���������ǽ��ڷ��յIJ�ȷ����ͨ��������ˮƽ��/CTE��
�����ͬ��������ȷ����ʧ��������������ӯ����
# IFRS 17 ��Լ�ֽ��� FCF ������
def ifrs17_fcf(policy, discount_rate_curve, risk_adjustment_pct):
    """
    ��Լ�ֽ��� = δ���ֽ�����ֵ + ���յ���
    Future Cash Flows = PV(outflows) - PV(inflows) + Risk Adjustment
    """
    pv_outflows = sum(cf * discount_factor(t, discount_rate_curve)
                      for t, cf in enumerate(policy.expected_claims))
    pv_inflows = sum(prem * discount_factor(t, discount_rate_curve)
                     for t, prem in enumerate(policy.expected_premiums))
    risk_adj = (pv_outflows - pv_inflows) * risk_adjustment_pct  # ����ֵ��5%-15%
    return pv_outflows - pv_inflows + risk_adj

6. Interest Rate Risk & ALM / ���ʷ������ʲ���ծƥ��|

2024��Ԥ�������µ������붨�۲���

����Ԥ���������۲�����Ʒ����
2024��9��1��ǰ��3.5%��ǰ������������ͳ����/�����
2024��9��1�պ���3.0%�����ʶ��ۻ�������������
2025��������3.0%���ȶ������ӷֺ측���ռ��ֺ���/������

����ѹ�����ԣ����������

def interest_rate_stress_test(policy, base_rate):
    scenarios = {
        "base": base_rate,
        "down_50bp": base_rate - 0.005,
        "down_100bp": base_rate - 0.010,
        "down_150bp": base_rate - 0.015,
    }
    results = {}
    for scenario, rate in scenarios.items():
        reserve = calculate_reserve(policy, rate)
        results[scenario] = {
            "reserve": reserve,
            "surplus_change": reserve - calculate_reserve(policy, base_rate)
        }
    return results

Reference Files / �ο��ļ�|

File / �ļ�Content / ����˵��
references/life_table_2025_usage.md������������(2025)ʹ��˵������CL1/CL2/CL3��������˵��
references/actuarial_pricing_formulas.md�����Ʒ���㶨�۹�ʽ���ܣ�������ʾ����
references/critical_illness_pricing.md�ؼ��ն���ר��������ӡ��ؼ������ʱ�
references/reserve_calculation.md׼�������ģ�壬������/������/����׼����
references/group_insurance_pricing.md���ն���ר���������ʱ�����ҵ����ϵ��

CL1/CL2/CL3�ؼ����������ʣ�2025�桤��ѡ��

CL1����������ҵ��һ��������/���������ã���

��������������(qx, ��)Ů��������(qx, ��)˵��
200.280.15�����ڣ������ʼ���
300.420.23����֮�꣬���Կ�ʼ����
401.120.67���꣬����������>Ů��1.7��
503.451.98֪��֮�꣬����������
608.925.14�����꣬���������ʽӽ�Ů��2��
7021.3413.87��ϡ֮�꣬�����С
8058.1242.56�ȳ�֮�꣬Ů��׷��

CL2����������ҵ���������ȫ/������ã���

��������������(qx, ��)Ů��������(qx, ��)˵��
200.250.13��CL1���ͣ������壩
300.380.20��ѡ�����
401.050.60������ѡ
503.201.85֪����ѡ
608.104.75������ѡ
7019.5012.80��ϡ��ѡ
8053.2039.10�ȳ���ѡ

CL3��������ҵ�����ר���������ã���

��������������(qx, ��)Ů��������(qx, ��)˵��
607.504.20���Ͻ���ȡ�ڿ�ʼ
6512.307.10���Կ�ʼ��������
7019.8012.50Ů�Լ�����
7531.2021.40��Ů�����խ
8048.5035.20������
8572.3056.10Ԥ������<5��
90103.4084.20�����䣬������>100��

������Դ���й�����ʦЭ�ᡶ�й���������ҵ������������2025����CL1/CL2/CL3��2025��10��29�շ�����2026��1��1����ǿ��ִ�С�

ʹ�ý�����

  • ��������/�������ղ�Ʒ �� ����ʹ�� CL1����������ߣ�������أ�
  • ��ȫ��/����գ������ϣ� �� ʹ�� CL2���������еȣ�
  • ר����ҵ���ϱ��� �� ����ʹ�� CL3����������ͣ����Ͻ����ѹ�����

GitHub: https://github.com/gechengling/insurance-actuarial-cn

核心工作流程(Dianjin融合版)

第一步:产品定位与定价目标(Dianjin精髓:市场洞察驱动)

定价维度核心问题决策输出
产品定位目标客群/保障范围/竞争格局产品形态、保障责任、费率策略
市场分析同类产品费率/市场份额/增长趋势定价竞争力分析、差异化策略
盈利目标目标利润率/投资收益率假设定价假设、利润测试标准
监管要求预定利率上限/偿付能力/准备金合规边界、监管风险点

定价目标输出

【定价目标书】
- 产品名称:[XXX]
- 目标客群:[年龄/性别/职业/收入]
- 定价假设:预定利率[X]%,死亡率表[CL1/CL2/CL3],费用率[X]%
- 目标利润率:[X]%(IRR口径)
- 监管要求:偿二代下资本占用[X]元/万元保费

第二步:发生率假设与经验分析(Dianjin精髓:大数据发生率)

2.1 发生率表选择框架

发生率类型推荐表适用产品监管认可
死亡率CL1/CL2/CL3(2025版)定寿/终身寿强制使用
重疾率CII1/CII2(2024版)重疾险/医疗险强制使用
医疗率MIR1/MIR2(2025版)医疗险/护理险推荐评审
年金率AFR1/AFR2(2025版)年金险/养老金强制使用
退保率SUR1/SUR2(经验)所有产品自行设定

2.2 经验发生率分析

【经验分析流程】
1. 数据清洗:剔除异常保单、修正录入错误
2. 经验提取:实际发生率 = f(年龄,性别,保单年度,渠道)
3. 趋势调整:医疗进步/疾病谱变化/退保行为变化
4. 置信区间:95%CI,最终发生率 = 经验±[X]%
5. 与行业对比:中国精算师协会经验表、同业数据

第三步:保费计算与利润测试(Dianjin精髓:动态利润分析)

3.1 纯保费计算公式

计算项目公式参数说明
纯保费(NP)NP = Σ CF_t × v^t × p_x+tCF=净现金流,v=贴现因子
费用附加L(oading) = α×NP + β×P + γα=固定%+β=变动%+γ=固定额
总保费(P)P = (NP + L) / (1 - 利润率%)含目标利润的总保费

3.2 利润测试模板

测试场景IRR(%)投资收益率(%)退保率(%)综合成本率(%)结论
基准场景[X][X][X][X]合格/不合格
悲观场景[X][X]-50bp[X]+20%[X]+200bp合格/不合格
乐观场景[X][X]+50bp[X]-20%[X]-200bp合格/不合格

IRR计算代码模板

# 定价IRR计算
import numpy as np

def calculate_irr(premiums, claims, expenses, reserves):
    cashflows = [premiums - claims - expenses - (reserves[t+1] - reserves[t]) 
                 for t in range(len(premiums))]
    return np.irr(cashflows)

# 示例:10年期缴重疾险
premiums = [10000] * 10  # 年缴1万
claims = [0, 5000, 15000, ...]  # 逐年理赔
irr = calculate_irr(premiums, claims, [500]*10, [0]*10)
print(f"定价IRR: {irr:.2%}")

第四步:准备金评估与偿付能力(Dianjin精髓:偿二代资本计量)

4.1 准备金计量框架

准备金类型计量基础监管公式资本占用
最优估计准备金(BEL)预期现金流现值BEL = Σ CF × v^t-
风险边际(RM)成本/置信度法RM = BEL × [X]%+
偿付能力准备金(SCR)Solvency II/C-ROSSSCR = max(BEL+RM, 监管要求)+

4.2 偿二代资本占用分析

【C-ROSS资本占用计算】
1. 保险风险资本(IC):死亡率/重疾率/退保率/费用风险
2. 市场风险资本(MC):利率/权益/房地产/汇率风险
3. 信用风险资本(CR):再保/债券/存款对手方风险
4. 操作风险资本(OR):固定+变动,按保费/准备金的[X]%
5. 总资本要求 = IC + MC + CR + OR

偿二代下资本占用:[X]元/万元保费

第五步:定价报告与监管备案(Dianjin精髓:标准化输出)

精算定价报告模板

# [产品名称] 精算定价报告

## 一、产品概述
[产品定位/目标客群/市场竞争分析]

## 二、定价假设
[生命表/重疾表/投资收益率/费用率/退保率]

## 三、保费计算
[纯保费/费用附加/总保费/IRR测试结果]

## 四、准备金与资本
[准备金计量/偿二代资本占用/IFRS17影响]

## 五、敏感性分析
[利率/死亡率/费用率/退保率±10%影响]

## 六、监管合规
[偿二代合规/准备金充足性/定价披露要求]

## 七、结论与建议
[定价结论/盈利能力/监管沟通策略]

合规约束与审计规则(Dianjin精髓)

  1. 监管合规:严格遵守《保险法》《精算工作规范》《偿二代规则》
  2. 数据质量:经验数据需清洗、校验,异常情况需说明
  3. 假设合理性:定价假设需有依据,不得为竞争恶性定价
  4. 准备金充足:准备金评估需保守,不得为利润释放不足提
  5. 审计留痕:所有计算过程、假设选择、模型选择需留痕可审计

测试用例(Dianjin精髓)

测试场景1:重疾险定价

  • 输入:30岁男性,20年缴,保额50万
  • 预期输出:年缴保费、IRR、准备金、资本占用
  • 通过标准:定价IRR>3.0%,资本占用<500元/万元

测试场景2:年金险准备金评估

  • 输入:终身年金,领取期20年,投资收益率4.0%
  • 预期输出:BEL、RM、SCR
  • 通过标准:SCR/准备金<15%

测试场景3:偿二代资本计算

  • 输入:重疾险产品组合,保费1亿
  • 预期输出:IC/MC/CR/OR分项,总资本要求
  • 通过标准:计算逻辑符合C-ROSS规则

关联技能(Dianjin精髓)

  • insurance-claims-intelligence:理赔数据→经验发生率更新
  • insurance-anti-fraud:欺诈识别→定价反欺诈因子
  • insurance-product-design:精算定价→产品设计协同
  • finance-ai-strategy:保险AI→精算AI战略协同

本技能融合阿里点金(Qwen Dianjin)金融AI精髓,专注经验发生率分析、动态利润测试、偿二代资本计量。

Questions people ask

Does this skill actually execute code or run models?
No. It is an educational reference; Python snippets are shown inline in markdown. No executable scripts, binaries, persistent storage, or network calls are included.
Which life table does it use?
The 4th Life Table (2025) with CL1 (life insurance), CL2 (annuity / health), and CL3 (speciality commercial pension) variants, released by the Chinese Actuarial Society on October 29, 2025 and mandatory from January 1, 2026.
Can its outputs be used directly for regulatory filings?
No. The document explicitly states it is not financial, legal, or insurance advice and that all outputs require human review before real-world application.

Related skills

Fetch raw ad creative, app, ranking, and revenue data from AdMapix as structured JSON.

by fly0pants

Trade crypto, manage a multi-chain wallet, and query an AI analyst from one CLI.

by lowesyang162 installs109 stars

Stores durable facts in a categorized, plain-markdown vault on disk, alongside your agent's built-in memory.

by Iván1 installs

Find why your productivity system keeps failing, then apply the smallest fix — capacity math, bottleneck routing, durable local notes.

by Iván1 installs

Adaptive web scraping in Python that bypasses anti-bot systems and scales from single requests to concurrent crawls.

by d4vinci399 installs28 stars

More from gechengling

Browse all skills

Map manual processes to autonomous AI pipelines — scored, blueprinted, and exportable to n8n or Make.

by gechengling28 installs1 stars

Compare and pick the right agent framework, with starter code and MCP server setup guidance.

by gechengling14 installs1 stars

Structured guidance for writing, auditing, and optimizing LLM prompts across frameworks and models.

by gechengling20 installs

A coach for AI-assisted app building — tool selection, prompt patterns, and workflows from idea to deploy.

by gechengling19 installs

Five-dimensional investment analysis with macro veto power, resonance grading, and position sizing.

by gechengling12 installs1 stars