SQL120 贷款情况

SQL120 贷款情况

withcity_loan_stats as (selectcity,sum(loan_amount) as total_loan_amount,round(sum(loan_amount) / count(distinct customer_id), 2) as average_loan_amount,count(distinct customer_id) as total_customersfromloan_applicationsjoin customers using (customer_id)group bycity),popular_loan_types as (selectcity,loan_type_name as most_applied_loan_typefrom(selectcity,loan_type_name,loan_type_id,row_number() over (partition bycityorder bycount(*) desc,loan_type_id asc) as rkfromloan_applicationsjoin customers using (customer_id)join (select distinctapplication_id,loan_type_name,loan_types.loan_type_idfromloan_typesjoin loan_application_types using (loan_type_id)) as t1 using (application_id)group bycity,loan_type_name,loan_type_id) as t2whererk = 1)
select*
fromcity_loan_statsjoin popular_loan_types using (city)
order bycity

SQL查询分析

这个查询的目的是分析各城市的贷款申请数据，主要关注三个方面的指标：贷款总额、客户平均贷款额，以及最受欢迎的贷款类型。

1. 整体结构

查询使用了两个CTE(Common Table Expressions)：

• city_loan_stats：计算城市级别的贷款汇总统计
• popular_loan_types：找出每个城市最受欢迎的贷款类型

最后将这两个CTE通过城市(city)关联，输出完整的分析结果。

2. 第一个CTE：city_loan_stats

SELECTcity,SUM(loan_amount) AS total_loan_amount,ROUND(SUM(loan_amount) / COUNT(DISTINCT customer_id), 2) AS avg_loan_per_customer,COUNT(DISTINCT customer_id) AS customer_count
FROMloan_applicationsJOIN customers USING (customer_id)
GROUP BYcity

​功能​：

• 计算每个城市的总贷款金额(total_loan_amount)
• 计算每个客户的平均贷款金额(avg_loan_per_customer)
• 统计每个城市的客户数量(customer_count)

​实现方式​：

1. 通过JOIN连接loan_applications和customers表
2. 按城市(city)分组
3. 使用聚合函数计算各项指标

3. 第二个CTE：popular_loan_types

SELECTcity,loan_type_name AS most_popular_loan_type
FROM(SELECTcity,loan_type_name,loan_type_id,ROW_NUMBER() OVER (PARTITION BY cityORDER BY COUNT(*) DESC, loan_type_id ASC) AS rankFROMloan_applicationsJOIN customers USING (customer_id)JOIN (SELECT DISTINCTapplication_id,loan_type_name,loan_types.loan_type_idFROMloan_typesJOIN loan_application_types USING (loan_type_id)) AS loan_type_data USING (application_id)GROUP BYcity,loan_type_name,loan_type_id) AS ranked_loans
WHERErank = 1

​功能​：

• 找出每个城市申请次数最多的贷款类型

​实现方式​：

1. 通过多层连接获取完整的贷款申请数据（包括城市和贷款类型信息）
2. 按城市和贷款类型分组统计申请次数
3. 使用窗口函数ROW_NUMBER()为每个城市的贷款类型按申请次数排序
   • PARTITION BY city：按城市分区
   • ORDER BY COUNT(*) DESC：按申请次数降序
   • loan_type_id ASC：次数相同时按loan_type_id升序（作为tie-breaker）
4. 筛选出排名第一的记录(rank = 1)

4. 最终查询

SELECTcls.city,cls.total_loan_amount,cls.avg_loan_per_customer,cls.customer_count,plt.most_popular_loan_type
FROMcity_loan_stats clsJOIN popular_loan_types plt USING (city)
ORDER BYcls.city

​功能​：

• 将两个CTE的结果按城市关联
• 输出完整的城市贷款分析报告
• 按城市名称排序

5. 业务价值

这个查询提供了以下业务洞察：

1. ​城市贷款规模​：通过总贷款金额了解哪些城市贷款业务量大
2. ​客户价值​：通过平均贷款金额了解不同城市客户的贷款规模
3. ​客户基础​：通过客户数量了解各城市市场渗透率
4. ​产品偏好​：了解不同城市客户最青睐的贷款产品类型