{"title":"On selecting a sample by probability proportional to size with second-order inclusion probabilities and without replacement","authors":"L. Mihályffy","doi":"10.20311/stat2016.k20.en083","DOIUrl":"https://doi.org/10.20311/stat2016.k20.en083","url":null,"abstract":"Given appropriate sets of first- and second-order inclusion probabilities, the author provides a method that results in samples including units and pairs of units of the universe with the probabilities specified in advance.","PeriodicalId":119089,"journal":{"name":"Hungarian Statistical Review","volume":"73 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115093521","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}
István Szabó, L. Zag, Irma F Takács, Balázs Kotosz, Dorina Csenki, E. Losoncz, J. Pap-Szekeres
{"title":"Quantile regression and an application: performance improvement of an emergency department in Eastern Europe","authors":"István Szabó, L. Zag, Irma F Takács, Balázs Kotosz, Dorina Csenki, E. Losoncz, J. Pap-Szekeres","doi":"10.35618/hsr2020.01.en060","DOIUrl":"https://doi.org/10.35618/hsr2020.01.en060","url":null,"abstract":"ED (emergency department) overcrowding is a problem faced by hospitals worldwide. Several studies have been performed to find solutions, but only few have proposed to decrease the length of stay by employing a radiologist in the ED. This study aims to improve emergency care in an Eastern European ED by measuring the parameters of crowding, introducing interventions based on the results, and evaluating their outcomes. As the length of stay is a typically skewed distribution variable, robust quantile regression is applied. The number of patients visiting the ED was measured from July 2014 to December 2015. The input, throughput and output parameters of ED crowding were evaluated throughout this period. The time intervals between the various stages of patient visits to the ED significantly decreased during the study period. The continuous measure-ment of ED process parameters is important to maintain time intervals within a specified range. Decreased process times between the pre- and post-intervention phases of the study were obtained by introducing several staff-centric changes. The presence of a dedicated radiologist in the ED has significantly decreased the turnaround times of imaging studies.","PeriodicalId":119089,"journal":{"name":"Hungarian Statistical Review","volume":"159 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114731429","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}
{"title":"Modelling MTPL insurance claim events: Can machine learning methods overperform the traditional GLM approach?","authors":"David Burka, László Kovács, László Szepesváry","doi":"10.35618/hsr2021.02.en034","DOIUrl":"https://doi.org/10.35618/hsr2021.02.en034","url":null,"abstract":"Pricing an insurance product covering motor third-party liability is a major challenge for actuaries. Comprehensive statistical modelling and modern computational power are necessary to solve this problem. The generalised linear and additive modelling approaches have been widely used by insurance companies for a long time. Modelling with modern machine learning methods has recently started, but applying them properly with relevant features is a great issue for pricing experts. This study analyses the claim-causing probability by fitting generalised linear modelling, generalised additive modelling, random forest, and neural network models. Several evaluation measures are used to compare these techniques. The best model is a mixture of the base methods. The authors’ hypothesis about the existence of significant interactions between feature variables is proved by the models. A simplified classification and visualisation is performed on the final model, which can support tariff applications later.","PeriodicalId":119089,"journal":{"name":"Hungarian Statistical Review","volume":"202 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133936577","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}
{"title":"A zero-truncated discrete Akash distribution with properties and applications","authors":"Simon Sium, R. Shanker","doi":"10.35618/hsr2020.02.en012","DOIUrl":"https://doi.org/10.35618/hsr2020.02.en012","url":null,"abstract":"This study proposes and examines a zero-truncated discrete Akash distribution and obtains its probability and moment-generating functions. Its moments and moments-based statistical constants, including coefficient of variation, skewness, kurtosis, and the index of dispersion, are also presented. The parameter estimation is discussed using both the method of moments and maximum likelihood. Applications of the distribution are explained through three examples of real datasets, which demonstrate that the zero-truncated discrete Akash distribution gives better fit than several zero-truncated discrete distributions.","PeriodicalId":119089,"journal":{"name":"Hungarian Statistical Review","volume":"121 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114518853","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}
{"title":"Providing new impetus to corporate well-being programmes: improving life expectancy through risk assessment","authors":"Gábor Vona","doi":"10.35618/hsr2021.01.en017","DOIUrl":"https://doi.org/10.35618/hsr2021.01.en017","url":null,"abstract":"Diseases of the circulatory system are leading causes of death, which compel stakeholders to lessen cardiovascular risks by utilising more effective prevention. These risks can be estimated based on gender, age, smoker status, systolic blood pressure, and total cholesterol. Artificial neural networks enable modelling of 10-year cardiovascular mortality rates. Understandable communica-tion of potential gains in life expectancy may enhance health consciousness through mitigating behavioural risks. The reproduction of death statistics requires the adjustment of the recommended probabilities for the occurrence of fatal cardiovascular events. This study deals with countries at high and low cardiovascular risk, selecting Hungary and the Czech Republic (high risk) and Austria (low risk). In Hungary, the gains in life expectancy are (43.4 – 36.2 =) 7.21 years for fe-males and (37.4 – 28.0 =) 9.4 years for males, both aged 40. These figures moderate to (21.2 – 15.6 =) 5.72 and (17.1 – 11.3 =) 5.8 years for elderly people aged 65, respectively. The Czech Republic represents an interim phase between the two other countries regarding ad-vancement in life expectancy, the respective gains exceed the Hungarian values: (45.8 – 37.8 =) 8.0, (39.7 – 29.7 =) 10.0, (23.0 – 16.6 =) 6.4, and (18.2 – 12.3 =) 5.9 years. In contrast, a 40-year-old woman may benefit from an additional (46.6 – 41.3 =) 5.3 years in Austria, while the corresponding accrual for men is (42.3 – 35.7 =) 6.6 years. On reaching 65 years, the increment is (23.4 – 19.3 =) 4.1 and (20.0 – 16.1 =) 3.9 years.","PeriodicalId":119089,"journal":{"name":"Hungarian Statistical Review","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121495204","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}
{"title":"Hungarians working abroad, non-residents working in Hungary (Difficulties in the statistical enumeration)","authors":"J. Lakatos","doi":"10.20311/stat2016.k20.en003","DOIUrl":"https://doi.org/10.20311/stat2016.k20.en003","url":null,"abstract":"Although the understanding of international migration is important for both the countries of origin and countries of destination, the tools of statistics are quite insufficient. The EU LFS (EU Labour Force Survey), which may be considered as mirror statistics and a target survey as well, is an important data source, nevertheless, methodological pitfalls may occur. Until the recent years, Hungary was hardly affected by the migration process of the EU, so its recording was less important. As a result of the 2008 crisis, however, the number of people working abroad has increased, and the growth was further intensified by lifting the restrictions in the Austrian and German labour market in the spring of 2011. For Hungary, the three main EU countries of destination are Austria, Germany and the United Kingdom. Labour movements to these countries include commuting, which can be traced by the Hungarian LFS. Migration processes have typical characteristics by the countries of destination, while the data sources available for estimating the size of migration and the relevance of their information are different also by countries. For recording the labour migration to Hungary, there are several administrative data sources, so it is not necessary to use the less reliable data of population surveys.","PeriodicalId":119089,"journal":{"name":"Hungarian Statistical Review","volume":"15 8","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114031674","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}