Pub Date : 2023-06-13DOI: 10.59170/stattrans-2023-034
G. Kalton
I thank the discussants for their kind remarks, for their insightful comments on the present state and future directions of the field, and for the many references they cite. Having no disagreements with them, I will confine my rejoinder to a few issues that their contributions have highlighted for me.
{"title":"Rejoinder","authors":"G. Kalton","doi":"10.59170/stattrans-2023-034","DOIUrl":"https://doi.org/10.59170/stattrans-2023-034","url":null,"abstract":"I thank the discussants for their kind remarks, for their insightful comments on\u0000 the present state and future directions of the field, and for the many references they\u0000 cite. Having no disagreements with them, I will confine my rejoinder to a few issues\u0000 that their contributions have highlighted for me.","PeriodicalId":37985,"journal":{"name":"Statistics in Transition","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47581144","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-13DOI: 10.59170/stattrans-2023-041
Jacek Białek
A wide variety of retailers (supermarkets, home electronics, Internet shops, etc.) provide scanner data containing information at the level of the barcode, e.g. the Global Trade Item Number (GTIN). As scanner data provide complete transaction information, we may use the expenditure shares of items as weightsfor calculating price indices at the lowest (elementary) level of data aggregation. The challenge here is the choice of the index formula which should be able to reduce chain drift bias and substitution bias. Multilateral index methods seem to be the best choice due to the dynamic character of scanner data. These indices work on a wholetime window and are transitive, which is key to the elimination of the chain drift effect. Following what is called an identity test, however, it may be expected that even when only prices return to their original values, the index becomes one. Unfortunately, the commonly used multilateral indices (GEKS, CCDI, GK, TPD, TDH) do not meet the identity test. The paper discusses the proposal of two multilateral indices and their weighted versions. On the one hand, the design of the proposed indices is based on the idea of the GEKS index. On the other hand, similarly to the Geary-Khamis method, it requires quality adjusting. It is shown that the proposed indices meet the identity test and most other tests. In an empirical and simulation study, these indices are compared with the SPQ index, which is relatively new and also meets the identity test. The analytical considerations as well as empirical studies confirm the high usefulness of the proposed indices.
{"title":"Quality adjusted GEKS-type indices for price comparisons based on scanner\u0000 data","authors":"Jacek Białek","doi":"10.59170/stattrans-2023-041","DOIUrl":"https://doi.org/10.59170/stattrans-2023-041","url":null,"abstract":"A wide variety of retailers (supermarkets, home electronics, Internet shops, etc.)\u0000 provide scanner data containing information at the level of the barcode, e.g. the Global\u0000 Trade Item Number (GTIN). As scanner data provide complete transaction information, we\u0000 may use the expenditure shares of items as weightsfor calculating price indices at the\u0000 lowest (elementary) level of data aggregation. The challenge here is the choice of the\u0000 index formula which should be able to reduce chain drift bias and substitution bias.\u0000 Multilateral index methods seem to be the best choice due to the dynamic character of\u0000 scanner data. These indices work on a wholetime window and are transitive, which is key\u0000 to the elimination of the chain drift effect. Following what is called an identity test,\u0000 however, it may be expected that even when only prices return to their original values,\u0000 the index becomes one. Unfortunately, the commonly used multilateral indices (GEKS,\u0000 CCDI, GK, TPD, TDH) do not meet the identity test. The paper discusses the proposal of\u0000 two multilateral indices and their weighted versions. On the one hand, the design of the\u0000 proposed indices is based on the idea of the GEKS index. On the other hand, similarly to\u0000 the Geary-Khamis method, it requires quality adjusting. It is shown that the proposed\u0000 indices meet the identity test and most other tests. In an empirical and simulation\u0000 study, these indices are compared with the SPQ index, which is relatively new and also\u0000 meets the identity test. The analytical considerations as well as empirical studies\u0000 confirm the high usefulness of the proposed indices.","PeriodicalId":37985,"journal":{"name":"Statistics in Transition","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44852663","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-13DOI: 10.59170/stattrans-2023-036
Bonginkosi D. Ndlovu, S. Melesse, T. Zewotir
icolaie et al. (2010) have advanced a vertical model as the latest continuous time competing risks model. The main objective of this article is to re-cast this model as a nonparametric model for analysis of discrete time competing risks data. Davis and Lawrance (1989) have advanced a cause-specific-hazard driven method for summarizing discrete time data nonparametrically. The secondary objective of this article is to compare the proposed model to this model. We pay particular attention to the estimates for the cause-specific-hazards and the cumulative incidence functions as well as their respective standard errors.
{"title":"A nonparametric analysis of discrete time competing risks data: a comparison of the\u0000 cause-specific-hazards approach and the vertical approach","authors":"Bonginkosi D. Ndlovu, S. Melesse, T. Zewotir","doi":"10.59170/stattrans-2023-036","DOIUrl":"https://doi.org/10.59170/stattrans-2023-036","url":null,"abstract":"icolaie et al. (2010) have advanced a vertical model as the latest continuous time\u0000 competing risks model. The main objective of this article is to re-cast this model as a\u0000 nonparametric model for analysis of discrete time competing risks data. Davis and\u0000 Lawrance (1989) have advanced a cause-specific-hazard driven method for summarizing\u0000 discrete time data nonparametrically. The secondary objective of this article is to\u0000 compare the proposed model to this model. We pay particular attention to the estimates\u0000 for the cause-specific-hazards and the cumulative incidence functions as well as their\u0000 respective standard errors.","PeriodicalId":37985,"journal":{"name":"Statistics in Transition","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43291856","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-13DOI: 10.59170/stattrans-2023-040
H. O. Ayhan
Survey statisticians have been dealing with the issues of nonresponse in sample surveys for many years. Due to the complex nature of the mechanism, so far it has not been easy to find a general solution to this problem. In this paper, several aspects of this topic will be elaborated on: the survey unit nonresponse bias has been examined alternatively by taking response amounts which are fixed initially and also by taking the response amounts as random variables. An overview of the components of the bias due to nonresponse will be performed. Nonresponse bias components are illustrated for each alternative approach and the amount of bias was computed for each case.
{"title":"Models for survey nonresponse and bias adjustment techniques","authors":"H. O. Ayhan","doi":"10.59170/stattrans-2023-040","DOIUrl":"https://doi.org/10.59170/stattrans-2023-040","url":null,"abstract":"Survey statisticians have been dealing with the issues of nonresponse in sample\u0000 surveys for many years. Due to the complex nature of the mechanism, so far it has not\u0000 been easy to find a general solution to this problem. In this paper, several aspects of\u0000 this topic will be elaborated on: the survey unit nonresponse bias has been examined\u0000 alternatively by taking response amounts which are fixed initially and also by taking\u0000 the response amounts as random variables. An overview of the components of the bias due\u0000 to nonresponse will be performed. Nonresponse bias components are illustrated for each\u0000 alternative approach and the amount of bias was computed for each case.","PeriodicalId":37985,"journal":{"name":"Statistics in Transition","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44317244","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-13DOI: 10.59170/stattrans-2023-030
D. Pfeffermann
I like to congratulate Professor Kalton for writing this very constructive article on probability versus nonprobability sampling. I learned a lot from reading it. In what follows, I add a few comments on this topic.
{"title":"Comments on „Probability vs. Nonprobability Sampling: From the Birth of Survey\u0000 Sampling to the Present Day” by Graham Kalton","authors":"D. Pfeffermann","doi":"10.59170/stattrans-2023-030","DOIUrl":"https://doi.org/10.59170/stattrans-2023-030","url":null,"abstract":"I like to congratulate Professor Kalton for writing this very constructive article\u0000 on probability versus nonprobability sampling. I learned a lot from reading it. In what\u0000 follows, I add a few comments on this topic.","PeriodicalId":37985,"journal":{"name":"Statistics in Transition","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48092827","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-13DOI: 10.59170/stattrans-2023-033
R. Münnich
Let me first thank Dr. Kalton for his amazing historical review of the development of survey sampling from its origin, contrasting purposive sampling, until now, where some elements of purposive sampling in terms of web or big data seem to supersede the well-elaborated theory of survey statistics. Shall the message be that we do not need any sampling courses at universities anymore, that official statistics should turn to modelling using data with unknown data generating processes, or actually even be substituted by (commercial) data krakens? Hardly so! Graham Kalton emphasises a modern thinking about the use of these new data sources which may also have some advantages and he urges future research on data integration methods using (very) different kinds of data while strongly taking quality aspects into account.
{"title":"Discussion of “Probability vs. Nonprobability Sampling: From the Birth of Survey\u0000 Sampling to the Present Day” by Graham Kalton","authors":"R. Münnich","doi":"10.59170/stattrans-2023-033","DOIUrl":"https://doi.org/10.59170/stattrans-2023-033","url":null,"abstract":"Let me first thank Dr. Kalton for his amazing historical review of the development\u0000 of survey sampling from its origin, contrasting purposive sampling, until now, where\u0000 some elements of purposive sampling in terms of web or big data seem to supersede the\u0000 well-elaborated theory of survey statistics. Shall the message be that we do not need\u0000 any sampling courses at universities anymore, that official statistics should turn to\u0000 modelling using data with unknown data generating processes, or actually even be\u0000 substituted by (commercial) data krakens? Hardly so! Graham Kalton emphasises a modern\u0000 thinking about the use of these new data sources which may also have some advantages and\u0000 he urges future research on data integration methods using (very) different kinds of\u0000 data while strongly taking quality aspects into account.","PeriodicalId":37985,"journal":{"name":"Statistics in Transition","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45069143","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-13DOI: 10.59170/stattrans-2023-044
A. Bhattacharjee, Rajashree Dey
In biomedical research, challenges to working with multiple events are often observed while dealing with time-to-event data. Studies on prolonged survival duration are prone to having numerous possibilities. In studies on prolonged survival, patients might die of other causes. Sometimes in the survival studies, patients experienced some events (e.g. cancer relapse) before dying within the study period. In this context, the semi-competing risks framework was found useful. Similarly, the prolonged duration of follow-up studies is also affected by censored observation, especially interval censoring, and right censoring. Some conventional approaches work with time-to-event data, like the Cox-proportional hazard model. However, the accelerated failure time (AFT) model is more effective than the Cox model because it overcomes the proportionality hazard assumption. We also observed covariates impacting the time-to-event data measured as the categorical format. No established method currently exists for fitting an AFT model that incorporates categorical covariates, multiple events, and censored observations simultaneously. This work is dedicated to overcoming the existing challenges by the applications of R programming and data illustration. We arrived at a conclusion that the developed methods are suitable to run and easy to implement in R software. The selection of covariates in the AFT model can be evaluated using model selection criteria such as the Deviance Information Criteria (DIC) and Log-pseudo marginal likelihood (LPML). Various extensions of the AFT model, such as AFT-DPM and AFT-LN, have been demonstrated. The final model was selected based on minimum DIC values and larger LPML values.
{"title":"Bayesian modelling for semi-competing risks data in the presence of\u0000 censoring","authors":"A. Bhattacharjee, Rajashree Dey","doi":"10.59170/stattrans-2023-044","DOIUrl":"https://doi.org/10.59170/stattrans-2023-044","url":null,"abstract":"In biomedical research, challenges to working with multiple events are often\u0000 observed while dealing with time-to-event data. Studies on prolonged survival duration\u0000 are prone to having numerous possibilities. In studies on prolonged survival, patients\u0000 might die of other causes. Sometimes in the survival studies, patients experienced some\u0000 events (e.g. cancer relapse) before dying within the study period. In this context, the\u0000 semi-competing risks framework was found useful. Similarly, the prolonged duration of\u0000 follow-up studies is also affected by censored observation, especially interval\u0000 censoring, and right censoring. Some conventional approaches work with time-to-event\u0000 data, like the Cox-proportional hazard model. However, the accelerated failure time\u0000 (AFT) model is more effective than the Cox model because it overcomes the\u0000 proportionality hazard assumption. We also observed covariates impacting the\u0000 time-to-event data measured as the categorical format. No established method currently\u0000 exists for fitting an AFT model that incorporates categorical covariates, multiple\u0000 events, and censored observations simultaneously. This work is dedicated to overcoming\u0000 the existing challenges by the applications of R programming and data illustration. We\u0000 arrived at a conclusion that the developed methods are suitable to run and easy to\u0000 implement in R software. The selection of covariates in the AFT model can be evaluated\u0000 using model selection criteria such as the Deviance Information Criteria (DIC) and\u0000 Log-pseudo marginal likelihood (LPML). Various extensions of the AFT model, such as\u0000 AFT-DPM and AFT-LN, have been demonstrated. The final model was selected based on\u0000 minimum DIC values and larger LPML values.","PeriodicalId":37985,"journal":{"name":"Statistics in Transition","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45306938","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-13DOI: 10.59170/stattrans-2023-035
J. Wesołowski
There is a continuing interplay between mathematics and survey methodology involving different branches of mathematics, not only probability. This interplay is quite obvious as regards the first of the two options: probability vs. non-probability sampling, as proposed and discussed in Kalton (2023). There, mathematics is represented by probability and mathematical statistics. However, sometimes connections between mathematics and survey methodology are less obvious, yet still crucial and intriguing. In this paper we refer to such an unexpected relation, namely between rotation sampling and Chebyshev polynomials. This connection, introduced in Kowalski and Wesołowski (2015), proved fundamental for the derivation of an explicit form of the recursion for the BLUE µˆt of the mean on each occasion t in repeated surveys based on a cascade rotation scheme. This general result was obtained under two basic assumptions: ASSUMPTION I and ASSUMPTION II, expressed in terms of the Chebyshev polynomials. Moreover, in that paper, it was conjectured that these two assumptions are always satisfied, so the derived form of recursion is universally valid. In this paper, we partially confirm this conjecture by showing that ASSUMPTION I is satisfied for rotation patterns with a single gap of an arbitrary size.
{"title":"Rotation schemes and Chebyshev polynomials","authors":"J. Wesołowski","doi":"10.59170/stattrans-2023-035","DOIUrl":"https://doi.org/10.59170/stattrans-2023-035","url":null,"abstract":"There is a continuing interplay between mathematics and survey methodology\u0000 involving different branches of mathematics, not only probability. This interplay is\u0000 quite obvious as regards the first of the two options: probability vs. non-probability\u0000 sampling, as proposed and discussed in Kalton (2023). There, mathematics is represented\u0000 by probability and mathematical statistics. However, sometimes connections between\u0000 mathematics and survey methodology are less obvious, yet still crucial and intriguing.\u0000 In this paper we refer to such an unexpected relation, namely between rotation sampling\u0000 and Chebyshev polynomials. This connection, introduced in Kowalski and Wesołowski\u0000 (2015), proved fundamental for the derivation of an explicit form of the recursion for\u0000 the BLUE µˆt of the mean on each occasion t in repeated surveys based on a cascade\u0000 rotation scheme. This general result was obtained under two basic assumptions:\u0000 ASSUMPTION I and ASSUMPTION II, expressed in terms of the Chebyshev polynomials.\u0000 Moreover, in that paper, it was conjectured that these two assumptions are always\u0000 satisfied, so the derived form of recursion is universally valid. In this paper, we\u0000 partially confirm this conjecture by showing that ASSUMPTION I is satisfied for rotation\u0000 patterns with a single gap of an arbitrary size.","PeriodicalId":37985,"journal":{"name":"Statistics in Transition","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42049990","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-13DOI: 10.59170/stattrans-2023-047
Nadeem Akhtar, S. Khan, Muhammad Amin, Akbar Ali Khan, Amjad Ali, Sadaf Manzoor
In this paper, the geometric distribution parameter is estimated under a type-I censoring scheme by means of the Bayesian estimation approach. The Beta and Kumaraswamy informative priors, as well as five loss functions are used for this purpose. Expressions of Bayes estimators and Bayes risks are derived under the Squared Error Loss Function (SELF), the Quadratic Loss Function (QLF), the Precautionary Loss Function (PLF), the Simple Asymmetric Precautionary Loss Function (SAPLF), and the DeGroot Loss Function (DLF) using the two aforementioned priors. The prior densities are obtained through prior predictive distributions. Simulation studies are carried out to make comparisons using Bayes risks. Finally, a real-life data example is used to verify the model’s efficiency.
{"title":"Bayesian estimation of a geometric distribution using informative priors based on a\u0000 Type-I censoring scheme","authors":"Nadeem Akhtar, S. Khan, Muhammad Amin, Akbar Ali Khan, Amjad Ali, Sadaf Manzoor","doi":"10.59170/stattrans-2023-047","DOIUrl":"https://doi.org/10.59170/stattrans-2023-047","url":null,"abstract":"In this paper, the geometric distribution parameter is estimated under a type-I\u0000 censoring scheme by means of the Bayesian estimation approach. The Beta and Kumaraswamy\u0000 informative priors, as well as five loss functions are used for this purpose.\u0000 Expressions of Bayes estimators and Bayes risks are derived under the Squared Error Loss\u0000 Function (SELF), the Quadratic Loss Function (QLF), the Precautionary Loss Function\u0000 (PLF), the Simple Asymmetric Precautionary Loss Function (SAPLF), and the DeGroot Loss\u0000 Function (DLF) using the two aforementioned priors. The prior densities are obtained\u0000 through prior predictive distributions. Simulation studies are carried out to make\u0000 comparisons using Bayes risks. Finally, a real-life data example is used to verify the\u0000 model’s efficiency.","PeriodicalId":37985,"journal":{"name":"Statistics in Transition","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45499345","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-13DOI: 10.59170/stattrans-2023-046
Wioletta Grzenda
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.
{"title":"Estimating the probability of leaving unemployment for older people in Poland using\u0000 survival models with censored data","authors":"Wioletta Grzenda","doi":"10.59170/stattrans-2023-046","DOIUrl":"https://doi.org/10.59170/stattrans-2023-046","url":null,"abstract":"Current demographic changes require greater participation of people aged 50 or\u0000 older in the labour market. Previous research shows that the chances of returning to\u0000 employment decrease with the length of the unemployment period. In the case of older\u0000 people who have not reached the statutory retirement age, these chances also depend on\u0000 the time they have left to retirement. Our study aims to assess the probability of\u0000 leaving unemployment for people aged 50-71 based on their characteristics and the length\u0000 of the unemployment period. We use data from the Labour Force Survey for 2019–2020. The\u0000 key factors determining employment status are identified using the proportional hazard\u0000 model. We take these factors into account and use the direct adjusted survival curve to\u0000 show how the probability of returning to work in Poland changes as people age. Due to\u0000 the fact that not many people take up employment around their retirement age, an\u0000 in-depth evaluation of the accuracy of predictions obtained via the models is crucial to\u0000 assess the results. Hence, in this paper, a time-dependent ROC curve is used. Our\u0000 results indicate that the key factor that influences the return to work after an\u0000 unemployment period in the case of older people in Poland is whether they reached the\u0000 age of 60. Other factors that proved important in this context are the sex and the\u0000 education level of older people.","PeriodicalId":37985,"journal":{"name":"Statistics in Transition","volume":"53 3","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41292523","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：