{"title":"使用带有删节数据的生存模型估计波兰老年人失业的可能性","authors":"Wioletta Grzenda","doi":"10.59170/stattrans-2023-046","DOIUrl":null,"url":null,"abstract":"Current demographic changes require greater participation of people aged 50 or\n older in the labour market. Previous research shows that the chances of returning to\n employment decrease with the length of the unemployment period. In the case of older\n people who have not reached the statutory retirement age, these chances also depend on\n the time they have left to retirement. Our study aims to assess the probability of\n leaving unemployment for people aged 50-71 based on their characteristics and the length\n of the unemployment period. We use data from the Labour Force Survey for 2019–2020. The\n key factors determining employment status are identified using the proportional hazard\n model. We take these factors into account and use the direct adjusted survival curve to\n show how the probability of returning to work in Poland changes as people age. Due to\n the fact that not many people take up employment around their retirement age, an\n in-depth evaluation of the accuracy of predictions obtained via the models is crucial to\n assess the results. Hence, in this paper, a time-dependent ROC curve is used. Our\n results indicate that the key factor that influences the return to work after an\n unemployment period in the case of older people in Poland is whether they reached the\n age of 60. Other factors that proved important in this context are the sex and the\n education level of older people.","PeriodicalId":37985,"journal":{"name":"Statistics in Transition","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2023-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Estimating the probability of leaving unemployment for older people in Poland using\\n survival models with censored data\",\"authors\":\"Wioletta Grzenda\",\"doi\":\"10.59170/stattrans-2023-046\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Current demographic changes require greater participation of people aged 50 or\\n older in the labour market. Previous research shows that the chances of returning to\\n employment decrease with the length of the unemployment period. In the case of older\\n people who have not reached the statutory retirement age, these chances also depend on\\n the time they have left to retirement. Our study aims to assess the probability of\\n leaving unemployment for people aged 50-71 based on their characteristics and the length\\n of the unemployment period. We use data from the Labour Force Survey for 2019–2020. The\\n key factors determining employment status are identified using the proportional hazard\\n model. We take these factors into account and use the direct adjusted survival curve to\\n show how the probability of returning to work in Poland changes as people age. Due to\\n the fact that not many people take up employment around their retirement age, an\\n in-depth evaluation of the accuracy of predictions obtained via the models is crucial to\\n assess the results. Hence, in this paper, a time-dependent ROC curve is used. Our\\n results indicate that the key factor that influences the return to work after an\\n unemployment period in the case of older people in Poland is whether they reached the\\n age of 60. Other factors that proved important in this context are the sex and the\\n education level of older people.\",\"PeriodicalId\":37985,\"journal\":{\"name\":\"Statistics in Transition\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2023-06-13\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Statistics in Transition\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.59170/stattrans-2023-046\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q4\",\"JCRName\":\"Mathematics\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Statistics in Transition","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.59170/stattrans-2023-046","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"Mathematics","Score":null,"Total":0}
Estimating the probability of leaving unemployment for older people in Poland using
survival models with censored data
Current demographic changes require greater participation of people aged 50 or
older in the labour market. Previous research shows that the chances of returning to
employment decrease with the length of the unemployment period. In the case of older
people who have not reached the statutory retirement age, these chances also depend on
the time they have left to retirement. Our study aims to assess the probability of
leaving unemployment for people aged 50-71 based on their characteristics and the length
of the unemployment period. We use data from the Labour Force Survey for 2019–2020. The
key factors determining employment status are identified using the proportional hazard
model. We take these factors into account and use the direct adjusted survival curve to
show how the probability of returning to work in Poland changes as people age. Due to
the fact that not many people take up employment around their retirement age, an
in-depth evaluation of the accuracy of predictions obtained via the models is crucial to
assess the results. Hence, in this paper, a time-dependent ROC curve is used. Our
results indicate that the key factor that influences the return to work after an
unemployment period in the case of older people in Poland is whether they reached the
age of 60. Other factors that proved important in this context are the sex and the
education level of older people.
期刊介绍:
Statistics in Transition (SiT) is an international journal published jointly by the Polish Statistical Association (PTS) and the Central Statistical Office of Poland (CSO/GUS), which sponsors this publication. Launched in 1993, it was issued twice a year until 2006; since then it appears - under a slightly changed title, Statistics in Transition new series - three times a year; and after 2013 as a regular quarterly journal." The journal provides a forum for exchange of ideas and experience amongst members of international community of statisticians, data producers and users, including researchers, teachers, policy makers and the general public. Its initially dominating focus on statistical issues pertinent to transition from centrally planned to a market-oriented economy has gradually been extended to embracing statistical problems related to development and modernization of the system of public (official) statistics, in general.