Insurance data for homeowners and motor insurance customers monitored over five years

Sample of 40,284 insurance private customers of a real portfolio of insurance policy holders in Spain with a motor insurance and a homeowners insurance contract. Customers are tracked from 2010 to 2014. Some customers do not renew their policies, so that they do not stay in the sample for five years. Data contain information on each customers, policies and yearly claims by type of contract. The file contains 122935 rows. Each customer is uniquely identified by an identification number, "PolID". Year of observation is recorded in column named “year” and it goes from 1 to 5. Records are presented ordered by years and by PolID within each year. Given that some policy holders do not renew the two policies in the insurance company, the number of customers per year decreases over time. The description of the data is the following: “gender" is 1 for male and 0 for female), "Age_client" is the age of the customer in years, "age_of_car_M" I s the number of years since the vehicle was bought by the customer, "Car_power_M" is the power of th vehicle, "Car_2ndDriver_M" equals 1 if the customer has informed the insurance company that a second occasional driver uses the vehicle, and 0 otherwise, "num_policiesC" is the total number of policies held by the same customer in the insurance company, "metro_code" equals 1 for urban or metropolitan and 0 for rural, "Policy_PaymentMethodA" equals 1 for annual payment and 0 for monthly payment in the motor policy, "Policy_PaymentMethodH" equals 1 for annual payment and 0 for monthly payment in the homeowners policy, "Insuredcapital_content_re" is the value of content in homeowners insurance , "Insuredcapital_continent_re" is the value of building in homeowners insurance, "appartment" equals 1 if the homeowners insurance correspond to an apartment and 0 otherwise "Client_Seniority" is the number of years that the customer has been in the company, "Retention" equals 1 if the policy is renewed and 0 otherwise,"NClaims1" is the number of claims in the motor insurance policy for the corresponding year, "NClaims2" is the number of claims in the motor insurance policy for the corresponding year, "Claims1" is the sum of claims cost in the motor insurance policy for the corresponding year,"Claims2" is the sum of claims cost in the homeowners insurance policy for the corresponding year, "Types" is equal to 1 when neither an auto nor a home claim, it is equal to 2 when the customer has an auto but not a home claim, it is equal to 3 when the customer does not have not an auto but a home claim and it is equal to 4 when both an auto and a home claim. All monetary units are expressed in Euros. In motor insurance, only claims at fault are considered.本数据集包含西班牙真实保险保单持有人组合的40284名私人客户样本，所有样本客户均同时投保机动车保险与家庭财产保险。研究跟踪周期为2010年至2014年，部分客户未续保时将不会完整覆盖五年的样本观测期。数据集涵盖每位客户、保单以及按保单类型划分的年度理赔相关信息。本文件共计122935条记录。每位客户均通过唯一标识号（PolID）进行识别，观测年份记录于名为"year"的列中，取值范围为1至5。记录按年份及同一年份内的PolID进行排序。由于部分投保人未在本保险公司续保两类保单，每年的客户数量随时间推移逐渐减少。 数据集各字段说明如下： "gender"：取值为1代表男性，0代表女性； "Age_client"：客户年龄（单位：年）； "age_of_car_M"：客户购置机动车后的使用年限； "Car_power_M"：机动车功率； "Car_2ndDriver_M"：若客户告知保险公司存在一名临时副驾驶员，则取值为1，否则为0； "num_policiesC"：该客户在本保险公司持有的总保单数量； "metro_code"：取值为1代表城市或大都会区域，0代表农村区域； "Policy_PaymentMethodA"：机动车保险的缴费方式，1为年度缴费，0为月度缴费； "Policy_PaymentMethodH"：家庭财产保险的缴费方式，1为年度缴费，0为月度缴费； "Insuredcapital_content_re"：家庭财产保险的室内财产投保保额； "Insuredcapital_continent_re"：家庭财产保险的建筑物投保保额； "appartment"：若家庭财产保险承保标的为公寓，则取值为1，否则为0； "Client_Seniority"：客户在本保险公司的服务年限； "Retention"：若保单续保时取值为1，否则为0； "NClaims1"：对应年度机动车保险保单的理赔次数； "NClaims2"：对应年度机动车保险保单的理赔次数； "Claims1"：对应年度机动车保险保单的理赔总支出； "Claims2"：对应年度家庭财产保险保单的理赔总支出； "Types"：取值为1代表既无机动车理赔也无家庭财产理赔；取值为2代表仅有机动车理赔但无家庭财产理赔；取值为3代表无机动车理赔但有家庭财产理赔；取值为4代表同时存在机动车与家庭财产理赔。 所有货币单位均以欧元计价。机动车保险仅统计有责理赔。