Meelis Käärik
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Meelis Käärik has a total of 32 co-authors. Here are the top co-authors by number of shared publications:
- Anne Selart
University of Tartu
3 shared publications Anne Selart
Institute of Mathematics and Statistics, University of Tartu, Tartu, Estonia
2 shared publicationsEne Käärik
Institute of Mathematical Statistics, University of Tartu, Estonia
2 shared publications- Ene Käärik
Institute of Mathematical Statistics, University of Tartu, J. Liivi 2, 50409 Tartu
2 shared publications - Kalev Pärna
University of Tartu
2 shared publications - Alexey Koloydenko
University of Nottingham
1 shared publication
Latest Publications
- Open Access
- 3 June 2025
- Published by University of Tartu in Acta et Commentationes Universitatis Tartuensis de Mathematica
- Vol. 29 (1) , 53-78
Lambert W normal distribution: a viable extension for skew-normal?
Several distributions and families of distributions, such as skew-normal and related distributions, are proposed to model skewed data. Lambert W transformation offers a flexible approach to introduce skewness to any random variable. In this paper, we compare the Lambert W normal distribution with the widely known skew-normal distribution and point out the similarities and differences, focusing mainly on risk-related characteristics and data fitting. While mathematically a different construction, one can think of Lambert W normal approach as an extension of the skew-normal to a more skewed situation.
- 3 June 2025
- Published by University of Tartu in Acta et Commentationes Universitatis Tartuensis de Mathematica
- Vol. 29 (1) , 53-78
Open AccessLambert W normal distribution: a viable extension for skew-normal?
Several distributions and families of distributions, such as skew-normal and related distributions, are proposed to model skewed data. Lambert W transformation offers a flexible approach to introduce skewness to any random variable. In this paper, we compare the Lambert W normal distribution with the widely known skew-normal distribution and point out the similarities and differences, focusing mainly on risk-related characteristics and data fitting. While mathematically a different construction, one can think of Lambert W normal approach as an extension of the skew-normal to a more skewed situation.
- Open Access
Lambert W Random Variables and Their Applications in Loss Modelling
Several distributions and families of distributions are proposed to model skewed data, e.g., with skew-normal and related distributions. Lambert W random variables offer an alternative approach in which, instead of constructing a new distribution, a certain transformation is proposed. Such an approach allows the construction of a Lambert W skewed version from any distribution. Here, we choose the Lambert W normal distribution as a natural starting point and include the Lambert W exponential distribution due to the simplicity and shape of the exponential distribution, which, after skewing, may produce a reasonably heavy tail for loss models. In the theoretical part, we focus on the mathematical properties of obtained distributions, including the range of skewness. In the practical part, the suitability of the corresponding Lambert W transformed distributions is evaluated on real insurance data. Finally, the results are compared with those obtained using common loss distributions.
Open AccessLambert W Random Variables and Their Applications in Loss Modelling
Several distributions and families of distributions are proposed to model skewed data, e.g., with skew-normal and related distributions. Lambert W random variables offer an alternative approach in which, instead of constructing a new distribution, a certain transformation is proposed. Such an approach allows the construction of a Lambert W skewed version from any distribution. Here, we choose the Lambert W normal distribution as a natural starting point and include the Lambert W exponential distribution due to the simplicity and shape of the exponential distribution, which, after skewing, may produce a reasonably heavy tail for loss models. In the theoretical part, we focus on the mathematical properties of obtained distributions, including the range of skewness. In the practical part, the suitability of the corresponding Lambert W transformed distributions is evaluated on real insurance data. Finally, the results are compared with those obtained using common loss distributions.
- Open Access
Analysis of follow-up data in large biobank cohorts: a review of methodology
This study focuses on key methodological challenges in genome-wide association studies (GWAS) of biobank data with time-to-event outcomes, analyzed using the Cox proportional hazards (CPH) model. We address four primary issues: left-truncation of the data, com-putational inefficiency of standard model-fitting algorithms, related-ness among individuals, and model misspecification. To manage left-truncation, the common practice is to use age as the timescale, with individuals entering the risk set at their age of recruitment. We assess how this choice of timescale influences bias and statistical power, under realistic GWAS conditions of varying effect sizes and censoring rates. In addition, to alleviate the computational burden typical in large-scale data, we propose and evaluate a two-step martingale residual (MR) approach for high-dimensional CPH modeling. Our results show that the timescale choice has minimal effect on accuracy for small hazard ratios, though using birth age as the timescale-ignoring recruitment age-yields the highest power for association detection. We find that relatedness, when ignored, does not substantially bias effect size estimates, while omitting key covariates introduces significant bias. The two-step MR approach proves to be computationally efficient, retaining power for detecting small effect sizes, making it suitable for large-scale association studies. However, when precise effect size estimates are critical, particularly for moderate or larger effect sizes, we recommend recalculating these estimates using the conventional CPH model, with careful attention to left-truncation and relatedness. These conclusions are drawn from simulations and illustrated with data from the Estonian Biobank cohort.
Open AccessAnalysis of follow-up data in large biobank cohorts: a review of methodology
This study focuses on key methodological challenges in genome-wide association studies (GWAS) of biobank data with time-to-event outcomes, analyzed using the Cox proportional hazards (CPH) model. We address four primary issues: left-truncation of the data, com-putational inefficiency of standard model-fitting algorithms, related-ness among individuals, and model misspecification. To manage left-truncation, the common practice is to use age as the timescale, with individuals entering the risk set at their age of recruitment. We assess how this choice of timescale influences bias and statistical power, under realistic GWAS conditions of varying effect sizes and censoring rates. In addition, to alleviate the computational burden typical in large-scale data, we propose and evaluate a two-step martingale residual (MR) approach for high-dimensional CPH modeling. Our results show that the timescale choice has minimal effect on accuracy for small hazard ratios, though using birth age as the timescale-ignoring recruitment age-yields the highest power for association detection. We find that relatedness, when ignored, does not substantially bias effect size estimates, while omitting key covariates introduces significant bias. The two-step MR approach proves to be computationally efficient, retaining power for detecting small effect sizes, making it suitable for large-scale association studies. However, when precise effect size estimates are critical, particularly for moderate or larger effect sizes, we recommend recalculating these estimates using the conventional CPH model, with careful attention to left-truncation and relatedness. These conclusions are drawn from simulations and illustrated with data from the Estonian Biobank cohort.
- Open Access
Multivariate Skew t-Distribution: Asymptotics for Parameter Estimators and Extension to Skew t-Copula
Symmetric elliptical distributions have been intensively used in data modeling and robustness studies. The area of applications was considerably widened after transforming elliptical distributions into the skew elliptical ones that preserve several good properties of the corresponding symmetric distributions and increase possibilities of data modeling. We consider three-parameter p-variate skew t-distribution where p-vector is the location parameter, is the positive definite scale parameter, p-vector is the skewness or shape parameter, and the number of degrees of freedom is fixed. Special attention is paid to the two-parameter distribution when that is useful for construction of the skew t-copula. Expressions of the parameters are presented through the moments and parameter estimates are found by the method of moments. Asymptotic normality is established for the estimators of and . Convergence to the asymptotic distributions is examined in simulation experiments.Open AccessMultivariate Skew t-Distribution: Asymptotics for Parameter Estimators and Extension to Skew t-Copula
Symmetric elliptical distributions have been intensively used in data modeling and robustness studies. The area of applications was considerably widened after transforming elliptical distributions into the skew elliptical ones that preserve several good properties of the corresponding symmetric distributions and increase possibilities of data modeling. We consider three-parameter p-variate skew t-distribution where p-vector is the location parameter, is the positive definite scale parameter, p-vector is the skewness or shape parameter, and the number of degrees of freedom is fixed. Special attention is paid to the two-parameter distribution when that is useful for construction of the skew t-copula. Expressions of the parameters are presented through the moments and parameter estimates are found by the method of moments. Asymptotic normality is established for the estimators of and . Convergence to the asymptotic distributions is examined in simulation experiments. - 28 August 2020
- Published by Taylor & Francis in The Clinical Neuropsychologist
- Vol. 35 (sup1) , S1-S20
WAIS-III measurement invariance: Data from Estonian standardization
A prerequisite of any psychological instrument used to compare individuals from different groups is measurement invariance (MI). It indicates that the test measures the same psychological construct...
- 28 August 2020
- Published by Taylor & Francis in The Clinical Neuropsychologist
- Vol. 35 (sup1) , S1-S20
WAIS-III measurement invariance: Data from Estonian standardization
A prerequisite of any psychological instrument used to compare individuals from different groups is measurement invariance (MI). It indicates that the test measures the same psychological construct...
- 24 December 2019
- Published by Taylor & Francis in Communications in Statistics - Simulation and Computation
- Vol. 51 (6) , 3013-3036
Estimating IBNR claim counts using different levels of data aggregation
Reserving problem in non-life insurance is an applied statistical problem with several computational aspects. In the literature, the focus has mainly been on aggregate reserving techniques and the chain-ladder method with its extensions has remained as the most widely applied claim reserving method. The classical chain-ladder method is regularly applied to annual data, but the question arises whether the reserve estimates based on more refined data outperform results obtained by annual data. We investigate whether and how much different data aggregation levels can improve the reserving process. To compare the performance of the classical chain-ladder method and its novel continuous extension, we conduct two simulation studies as well as a case study with an insurance data and use both models to estimate the IBNR claim count estimates on different levels of data aggregation. The results demonstrate that the continuous approach of the chain-ladder method provides in general a minor improvement in the IBNR predictions but proves to hold its predictive power on different types of data. This study highlights data aggregation levels in which the estimates of classical chain-ladder model outperforms the estimates obtained by the use of annual data, providing more insights for establishing accurate loss reserves.
- 24 December 2019
- Published by Taylor & Francis in Communications in Statistics - Simulation and Computation
- Vol. 51 (6) , 3013-3036
Estimating IBNR claim counts using different levels of data aggregation
Reserving problem in non-life insurance is an applied statistical problem with several computational aspects. In the literature, the focus has mainly been on aggregate reserving techniques and the chain-ladder method with its extensions has remained as the most widely applied claim reserving method. The classical chain-ladder method is regularly applied to annual data, but the question arises whether the reserve estimates based on more refined data outperform results obtained by annual data. We investigate whether and how much different data aggregation levels can improve the reserving process. To compare the performance of the classical chain-ladder method and its novel continuous extension, we conduct two simulation studies as well as a case study with an insurance data and use both models to estimate the IBNR claim count estimates on different levels of data aggregation. The results demonstrate that the continuous approach of the chain-ladder method provides in general a minor improvement in the IBNR predictions but proves to hold its predictive power on different types of data. This study highlights data aggregation levels in which the estimates of classical chain-ladder model outperforms the estimates obtained by the use of annual data, providing more insights for establishing accurate loss reserves.
- 23 October 2017
- Published by Taylor & Francis in Communications in Statistics - Theory and Methods
- Vol. 47 (15) , 3640-3655
Asymptotic normality of estimators for parameters of a multivariate skew-normal distribution
- 23 October 2017
- Published by Taylor & Francis in Communications in Statistics - Theory and Methods
- Vol. 47 (15) , 3640-3655
Asymptotic normality of estimators for parameters of a multivariate skew-normal distribution
- 3 July 2017
- Published by University of Tartu in Acta et Commentationes Universitatis Tartuensis de Mathematica
- Vol. 21 (1) , 51-63
On estimation of insurance risk parameters by combining local regression and distribution fitting ideas
The problem of premium estimation is an essential part of the insurance mathematics. Often the problem is divided into two parts: estimation of claim number (or frequency) and the estimation of individual claim amounts (severities). In this paper, we will focus on the former. More precisely, we are looking for certain semiparametric dynamic regression type model to avoid the "price shock" issue of static classication. We apply locally the regression method, use local maximum likelihood estimation for the parameters of the model and cross-validation techniques to determine the optimal size of a neighborhood. A case study with real vehicle casco insurance dataset is included, the results obtained by proposed method are compared by the ones obtained by global regression and the classification and regression trees (C&RT) approach.
- 3 July 2017
- Published by University of Tartu in Acta et Commentationes Universitatis Tartuensis de Mathematica
- Vol. 21 (1) , 51-63
On estimation of insurance risk parameters by combining local regression and distribution fitting ideas
The problem of premium estimation is an essential part of the insurance mathematics. Often the problem is divided into two parts: estimation of claim number (or frequency) and the estimation of individual claim amounts (severities). In this paper, we will focus on the former. More precisely, we are looking for certain semiparametric dynamic regression type model to avoid the "price shock" issue of static classication. We apply locally the regression method, use local maximum likelihood estimation for the parameters of the model and cross-validation techniques to determine the optimal size of a neighborhood. A case study with real vehicle casco insurance dataset is included, the results obtained by proposed method are compared by the ones obtained by global regression and the classification and regression trees (C&RT) approach.
- Open Access
On Comparison of Stochastic Reserving Methods with Bootstrapping
We consider the well-known stochastic reserve estimation methods on the basis of generalized linear models, such as the (over-dispersed) Poisson model, the gamma model and the log-normal model. For the likely variability of the claims reserve, bootstrap method is considered. In the bootstrapping framework, we discuss the choice of residuals, namely the Pearson residuals, the deviance residuals and the Anscombe residuals. In addition, several possible residual adjustments are discussed and compared in a case study. We carry out a practical implementation and comparison of methods using real-life insurance data to estimate reserves and their prediction errors. We propose to consider proper scoring rules for model validation, and the assessments will be drawn from an extensive case study.
Open AccessOn Comparison of Stochastic Reserving Methods with Bootstrapping
We consider the well-known stochastic reserve estimation methods on the basis of generalized linear models, such as the (over-dispersed) Poisson model, the gamma model and the log-normal model. For the likely variability of the claims reserve, bootstrap method is considered. In the bootstrapping framework, we discuss the choice of residuals, namely the Pearson residuals, the deviance residuals and the Anscombe residuals. In addition, several possible residual adjustments are discussed and compared in a case study. We carry out a practical implementation and comparison of methods using real-life insurance data to estimate reserves and their prediction errors. We propose to consider proper scoring rules for model validation, and the assessments will be drawn from an extensive case study.