765 Hits in 1.1 sec

Goodness-of-fit Testing in Linear Regression Models [article]

Rok Blagus, Jakob Peterlin, Janez Stare
2019 arXiv   pre-print
Model checking plays an important role in linear regression as model misspecification seriously affects the validity and efficiency of regression analysis. In practice, model checking is often performed by subjectively evaluating the plot of the model's residuals. This approach is objectified by constructing a random process from the model's residuals, however due to a very complex covariance function obtaining the exact distribution of the test statistic is intractable. Several solutions to
more » ... rcome this have been proposed, however the simulation and bootstrap based approaches are only asymptotically valid and can, with a limited sample size, yield tests which have inappropriate size. We therefore propose to estimate the null distribution by using permutations. We show, under some mild assumptions, that with homoscedastic random errors this yields consistent tests under the null and the alternative hypotheses. Small sample properties of the proposed tests are studied in an extensive Monte Carlo simulation study, where it is demonstrated that the proposed tests attain correct size, even with strongly non-normal random errors and a very small sample size, while being as powerful as the other available alternatives. The results are also illustrated on some real data examples.
arXiv:1911.07522v1 fatcat:vthuku7hwzaj3kntqokjngtb24

The Management of Complaints in Slovene Public Administration

Janez Stare, Barbara Gruden, Polonca Kovač
2008 Organizacija  
The Slovene public administration is part of the broader social system, therefore it must be responsive and proactive. The instrument of complaint in the administrative procedure, and wider in the context of the entire administrative management, is very helpful to public administration for tracking social changes and should therefore be seen as a form of constructive criticism. On the basis of user complaints regarding its services or the parties to the administrative procedure, the public
more » ... istration must learn permanently and, as learning organisation, must incorporate its findings into future practice. The starting point of this contribution is the problems of the complaint in the narrow sense -customer dissatisfaction with the functioning of the public administration or arising from the fact that one of the characteristics of the administrative procedure or other services is the direct contact between the customer and the provider of the service. With this approach, the public administration will develop part of an integral system of quality and excellence that underlines the meaning of satisfaction of (all) the users of public services. Upravljanje prito'b v slovenski javni upravi Slovenska javna uprava je del širšega dru'benega sistema. Zato mora delovati odzivno in celo proaktivno. Instrument prito'be v upravnem postopku in širše v okviru celotnega upravnega poslovanja je javni upravi pri sledenju dru'benim spremembam v pomo~ in bi ga tako morala jemati kot obliko konstruktivne kritike. Javna uprava se mora na podlagi prito'b uporabnikov njenih storitev oz. strank v upravnih postopkih u~iti in kot u~e~a organizacija nauke vgraditi v bodo~o prakso. V prispevku izhajamo iz tistih problemov prito'be v o'jem smislu kot nezadovoljstva stranke, ki se nanašajo na delovanje javne uprave oz. izhajajo iz dejstva, da je ena izmed zna~ilnosti upravnega postopka oz. druge storitve prav neposreden stik med stranko in izvajalcem storitve. S tovrstnim pristopom bo javna uprava razvila del celovitega sistema kakovosti oz. odli~nosti, ki poudarja pomen zadovoljstva (vseh) uporabnikov javnih storitev.
doi:10.2478/v10051-008-0003-9 fatcat:cbl4tl26afdzvev7vgkondu63i

Explained randomness in proportional hazards models

John O'Quigley, Ronghui Xu, Janez Stare
2005 Statistics in Medicine  
A coe cient of explained randomness, analogous to explained variation but for non-linear models, was presented by Kent. The construct hinges upon the notion of Kullback-Leibler information gain. Kent and O'Quigley developed these ideas, obtaining simple, multiple and partial coe cients for the situation of proportional hazards regression. Their approach was based upon the idea of transforming a general proportional hazards model to a speciÿc one of Weibull form. Xu and O'Quigley developed a
more » ... direct approach, more in harmony with the semi-parametric nature of the proportional hazards model thereby simplifying inference and allowing, for instance, the use of time dependent covariates. A potential drawback to the coe cient of Xu and O'Quigley is its interpretation as explained randomness in the covariate given time. An investigator might feel that the interpretation of the Kent and O'Quigley coe cient, as a proportion of explained randomness of time given the covariate, is preferable. One purpose of this note is to indicate that, under an independent censoring assumption, the two population coe cients coincide. Thus the simpler inferential setting for Xu and O'Quigley can also be applied to the coe cient of Kent and O'Quigley. Our second purpose is to point out that a sample-based coe cient in common use in the SAS statistical package can be interpreted as an estimate of explained randomness when there is no censoring. When there is censoring the SAS coe cient would not seem satisfactory in that its population counterpart depends on an independent censoring mechanism. However there is a quick ÿx and we argue in favour of its use.
doi:10.1002/sim.1946 pmid:15532086 fatcat:wf4esmubf5bg7b2dymibgtq2ia

Making relative survival analysis relatively easy

Maja Pohar, Janez Stare
2007 Computers in Biology and Medicine  
In survival analysis we are interested in time from the beginning of an observation until certain event (death, relapse, etc.). We assume that the final event is well defined, so that we are never in doubt whether the final event has occurred or not. In practice this is not always true. If we are interested in cause-specific deaths, then it may sometimes be difficult or even impossible to establish the cause of death, or there may be different causes of death, making it impossible to assign
more » ... h to just one cause. Suicides of terminal cancer patients are a typical example. In such cases, standard survival techniques cannot be used for estimation of mortality due to a certain cause. The cure to the problem are relative survival techniques which compare the survival experience in a study cohort to the one expected should they follow the background population mortality rates. This enables the estimation of the proportion of deaths due to a certain cause. In this paper, we briefly review some of the techniques to model relative survival, and outline a new fitting method for the additive model, which solves the problem of dependency of the parameter estimation on the assumption about the baseline excess hazard. We then direct the reader's attention to our R package relsurv that provides functions for easy and flexible fitting of all the commonly used relative survival regression models. The basic features of the package have been described in detail elsewhere, but here we additionally explain the usage of the new fitting method and the interface for using population mortality data freely available on the Internet. The combination of the package and the data sets provides a powerful informational tool in the hands of a skilled statistician/informatician. Motivation If a person with an incurable disease commits suicide, the cause of death written in the death certificate will be suicide. And if there were many such cases, the mortality statistics would show much lower proportion of deaths due to the disease in question than it really should. And while suicides are just an obvious, more or less hypothetical, example, it is less well known that it is often difficult or even impossible to select among different possible causes of death or assign a certain cause at all. People with a certain condition (e.g. diabetes, high blood pressure, etc.) may die of natural causes, but it is quite possible that they would have lived longer without that condition. In such cases we need methods of relative survival to estimate the proportion of people dying due to a certain cause. These methods are widely used in cancer registries, but almost never in other areas of medicine. The goal of this paper is to bring the methods of relative survival to a wider, possibly less statistical, audience, by giving an overview of the existing methods, outline some new methods, describe new possibilities of acquiring population data, and present a software package that includes functions for easy and flexible fitting of all the commonly used relative survival regression models.
doi:10.1016/j.compbiomed.2007.04.010 pmid:17582396 fatcat:m34sq4q3w5f5rmgajzemh5jip4

Accuracy of point predictions in survival analysis

Robin Henderson, Margaret Jones, Janez Stare
2001 Statistics in Medicine  
Survival time prediction is important in many applications, particularly for patients diagnosed with terminal diseases. A measure of prediction error taken from the medical literature is advocated as a practicable method of quantifying reliability of point predictions. Optimum predictions are derived for familiar survival models and the accuracy of these predictions is investigated. We argue that poor predictive capability is inherent to standard survival models with realistic parameter values.
more » ... A lung cancer example is used to illustrate di culties in prediction in practice.
doi:10.1002/sim.913 pmid:11590634 fatcat:knlxfwvmendh3gqmdzi7rsxpqe

Goodness of fit of relative survival models

Janez Stare, Maja Pohar, Robin Henderson
2005 Statistics in Medicine  
any model which relates observed survival to background, including also the multiplicative model class (Buckley 1984 , Andersen et al 1985 , Andersen and Vaeth 1989 and the transformation approach (Stare  ...  In turn, working from the ideas of O'Quigley (2003) and O'Quigley and Stare (2003) , we form a cumulative sum process: B n (β, k n ) := 1 √ n k i=1 R i (β), k = 1, . . . , n; B(β, 0) := 0 Here n denotes  ... 
doi:10.1002/sim.2414 pmid:16320279 fatcat:irmlsnjg6zajtaid2abyhlmcx4

A Measure of Explained Variation for Event History Data

Janez Stare, Maja Pohar Perme, Robin Henderson
2010 Biometrics  
coefficient of determination R 2 in linear regression, the search for an analogous measure in survival analysis has been pursued for almost 30 years (eg Harrell et al 1982 , Korn and Simon 1990 , Schemper and Stare  ... 
doi:10.1111/j.1541-0420.2010.01526.x pmid:21155749 fatcat:m6wm34pkfzd4rl5gp3xmujxtuu

Explained variation of excess hazard models

Camille Maringe, Maja Pohar Perme, Janez Stare, Bernard Rachet
2018 Statistics in Medicine  
The weighting initially proposed in Stare et al takes good account of random censoring through time.  ...  We believe the null model presented and used in Stare et al 13 in conjunction with our weighing remains the most relevant approach for the adaptation of the original RE to excess hazard models.  ... 
doi:10.1002/sim.7645 pmid:29633343 pmcid:PMC6001643 fatcat:cuzurh7m5jb7te4vnybivketie

Bad luck of cancer – or misinterpreted statistics?

Janez Stare
2017 Metodološki zvezki. Advances in methodology and statistics  
A paper in Science (January 2015) claimed that the majority, 65% to be precise, of cancers is due to bad luck, so non-preventable. In this paper we show that the analyses, presented in the paper, give absolutely no grounds to make such a claim. Some of the arguments have in the meantime appeared elsewhere, but some have not. We also show that the authors' assumptions and their data can only support a claim of no more than 5% of cancers being random.
doi:10.51936/tvcb8684 fatcat:sk2w6hlhybd2jpqr4wdww6snna

Does a Voluntary Job Change Affect Work-Family Balance?

Nina Tomazevic, Tatjana Kozjek, Janez Stare
2015 International Business Research  
The question of work-family balance (WFB) is an area where increasing attention is being paid nowadays since it relates to the health, satisfaction, personal development of employees, to the care for their children, youths and elder family members. The purpose of the paper is to explore the connection between the voluntary job change and the potential improvement of the work-family balance since it was assumed that changing the job voluntarily might bring to a higher WFB as employees could
more » ... fy their needs more effectively in a new working environment, chosen by their own will. Data were collected using a combination of CAWI (computer assisted web interview) and pen and pencil methods. Employees from various companies and public sector organizations from Slovenia were invited to participate in the study. Hierarchical cluster analysis was performed to identify clusters of family-work balancers, i.e. of those, who help with balancing employees" work and family domains. Kruskal-Wallis test and Mann-Whitney U test were used to explore the correlation between the decision about the voluntary job change and the satisfaction with the work-family balance. The research has shown that the assessment of WFB does not depend on a voluntary job change. The exceptions are the group of employees with children (in a positive direction) and the group of employees without children and no help in the household (in a negative direction). The employees do not use a job change as a tool for increasing WFB. This is probably also due to the closed culture of Slovenians. The managers who do care about their employees and who do not wish to lose their best employees have therefore many possibilities to choose the proper coping strategies to help their employees balance their work and family lives.
doi:10.5539/ibr.v8n2p1 fatcat:idejp4pd55btxkbpwf445eid5q

Odds ratio, hazard ratio and relative risk

Janez Stare, Delphine Maucort-Boulch
2016 Metodološki zvezki. Advances in methodology and statistics  
Odds ratio (OR) is a statistic commonly encountered in professional or scientific medical literature. Most readers perceive it as relative risk (RR), although most of them do not know why that would be true. But since such perception is mostly correct, there is nothing (or almost nothing) wrong with that. It is nevertheless useful to be reminded now and then what is the relation between the relative risk and the odds ratio, and when by equating the two statistics we are sometimes forcing OR to
more » ... e something it is not. Another statistic, which is often also perceived as a relative risk, is the hazard ratio (HR). We encounter it, for example, when we fit the Cox model to survival data. Under proportional hazards it is probably "natural" to think in the following way: if the probability of death in one group is at every time point k-times as high as the probability of death in another group, then the relative risk must be k, regardless of where in time we are. This could be hardly further from the truth and in this paper we try to dispense with this blunder.
doi:10.51936/uwah2960 fatcat:iat3phhmpzdndhljldoa6onx7y

Introducing Social Responsibility in Local Government Bodies and the Golden Thread Project

Bernarda Miklavc, Janez Stare
2014 Management  
doi:10.7595/management.fon.2014.0017 fatcat:ahdbkabdufdebnunks7kr72v5e

The Measurement Model of Professional Operation of State Administration

Milan Rman, Marjan Brezovšek, Janez Stare
2020 Central European Public Administration Review  
Stare Descriptive statistics of variables in a model From variables, indicated in Table 3 , we created new variables as the averages of all retained variables.  ...  coefficient, it shows reliability of constructs, by which an exception is only the "autonomy" construct, however on the other side of composite reliability is relatively high Milan Rman, Marjan Brezovšek, Janez  ... 
doi:10.17573/cepar.2020.2.02 fatcat:psfn5tqhsbadne4466gr6laa2a

E-Government and Cost-Effectiveness: E-Taxation in Slovenia

Mitja DEČMAN, Janez STARE, Maja KLUN
2010 Transylvanian Review of Administrative Sciences  
Improving government performance has been a key goal of regulatory reforms. Cost effectiveness, co-operation and bottom-up approach, flexibility, dynamics, responsiveness are important attributes of reforms. Improvements in regulatory performance include targets to reduce administrative burdens, and information technology is one of the tools that could be used for achieving that goal. The paper presents the use of information communication technology (ICT) in government procedures and its
more » ... on the cost-efficiency of government. The research focuses on investment in ICT to simplify tax procedures in Slovenia. Results show that ICT expenditure is higher than cost savings for tax administration and taxpayers. Nevertheless, several non-financial benefits are also important and should be considered.
doaj:464bb16db7fc4becb63ec3bff950b476 fatcat:fkpmo6b2uraj5h4oa3tx57fbhu

Training Needs Assessment for Leaders in Nursing Based on Comparison of Competency Models

Andreja Kvas, Janko Seljak, Janez Stare
2014 Organizacija  
When we define normative values of individual (required) competencies within the context of a competency model, we get a competency profile (Stare, 2009; .  ...  The "Competency model for leaders in state administration" (Stare 2009; Kvas, 2013) covers 77 behaviours or actions that are divided into seven groups of competencies.  ... 
doi:10.2478/orga-2014-0006 fatcat:gwmkyvcznneinmfqozbub57vzi
« Previous Showing results 1 — 15 out of 765 results