|
|
Statistical inference in varying coefficient models
Splítek, Martin ; Maciak, Matúš (advisor) ; Pešta, Michal (referee)
Tato práce se zabývá modely s promìnlivými koe cienty se za- mìøením na statistickou inferenci. Hlavní my¹lenkou tìchto modelù je pou¾ití regresních koe cientù, mìnících se v závislosti na nìjakém modi kátoru vlivu, namísto konstantních koe cientù klasické lineární regrese. Nejprve si de nujeme tyto modely a jejich odhadové procedury, kterých bylo doposud publikováno nì- kolik variant. K odhadu se pou¾ívá lokální regrese nebo rùzné druhy splajnù { vyhlazovací, polynomiální èi penalizované. Od metody odhadu se následnì od- víjí i daná statistická inference, ke které uvedeme odvozené vychýlení, rozptyl, asymptotickou normalitu, kon denèní pásma a testování hypotéz. Hlavním cílem na¹í práce je kompaktnì shrnout vybrané metody a jejich inferenci. Na závìr je navr¾ena proceduru pro výbìr promìnných.
|
|
|
Hurdle models in non-life insurance
Tian, Cheng ; Pešta, Michal (advisor) ; Branda, Martin (referee)
A number of articles only present hurdle models for count data. we are motivated to present hurdle models for semi-continuous data. Because semi- continuous data is also commonly seen in non-life insurance. The thesis deals with the parameterization of various hurdle models for semi-continuous data besides for count data in non-life insurance. Two components of a hurdle model are modeled separately. A hurdle component is modeled by a logistic regression. For a semi-continuous data, a continuous component is modeled by several various regressions. Parameters of each component are estimated through maximum likelihood estimation. Model selection is mentioned before theoretical approaches are applied on the vehicle insurance data. Finally, we get some predicted values based on the fitted models. The prediction gives insurance companies a general idea on setting premium but not accurate. 1
|
|
|
Loss reserving for individual claim-by-claim data
Bednárik, Vojtěch ; Pešta, Michal (advisor) ; Hurt, Jan (referee)
This thesis covers stochastic claims reserving in non-life insurance based on individual claims developments. Summarized theoretical methods are applied on data from Czech Insurers' Bureau for educational purposes. The problem of estimation is divided into four parts: oc- curence process generating claims, delay of notification, times between events and payments. Each part is estimated separately based on maximum likelihood theory and final estimates allow us to obtain an estimate of future liabilities distribution. The results are very promis- ing and we believe this method is worth of a further research. Contribution of this work is more rigorous theoretical part and application on data from the Czech market with some new ideas in practical part and simulation. 1
|
|
|
Maximum likelihood theory for not i.i.d. observations
Kielkowská, Eva ; Omelka, Marek (advisor) ; Pešta, Michal (referee)
Maximum likelihood approach for independent but not identically distributed observations is studied. In the first part of the thesis, conditions for consistency and asymptotic normality of the maximum likelihood estimates for this case are stated. Uniform integrability has a major role in proving the desired properties. K-sample problem serves as an example for using the described method. The second part is focused on estimates obtained by minimizing convex functions. Convexity is a key for showing the consistency and asymptotic normality of the estimates in this case. The results can be used for maximum likelihood when observations with logconcave densities are involved. Finally, normal linear model, logistic regression and Poisson regression examples are provided to present the application of the method.
|
|
|
Problem of the nearest correlation matrix
Sotáková, Martina ; Pešta, Michal (advisor) ; Maciak, Matúš (referee)
This work deals with the problem of finding the correlation matrix closest to the given symetric matrix, the distance of which is measured considering the Frobenius norm. The theoretical part of the thesis describes a method used for finding the solution to this problem based on the dual approach and application of Newton method. The method is further modified for other cases. In the practical part we apply the theory to simple math problems.
|
|
|
Bonus - Malus System with Deductibles
Kubát, Petr ; Mazurová, Lucie (advisor) ; Pešta, Michal (referee)
This thesis deals with the option of substitution of malus surcharge on pre- mium in a classical bonus - malus system with deductible. Firstly, we clarify the basic principles of bonus - malus systems, then we show how to model the expec- ted claim amount of the insureds based on their characteristics and we explain how to correctly select values of premium discounts and surcharges in the classes of bonus - malus systems. Next we clarify the concept of deductible and introduce the technique of its application on these systems. Finally we show the practical application of deductible on two models of bonus - malus systems and we evaluate and compare the results. 1
|
|
|
Mixed Poisson models for claim counts
Hauptfleisch, Filip ; Pešta, Michal (advisor) ; Hendrych, Radek (referee)
The thesis summarizes the theory of mixed Poisson models. Poisson distri- bution is one of the popular distributions in modelling count data, but its use is limited because it requires equidispersion. Because of this we introduce both con- tinuous and finite mixtures. From continuous mixtures the main representative is the negative binomial model, which arises as Poisson Gamma mixture, while from discrete models we deal mainly with zero-inflated models and hurdle models. For these models we use the maximum likelihood estimates of their parameters. In the end we apply these models to fit automobile insurance data from Australia, where we use MLE to fit Poisson regression, negative binomial regression and Poisson hurdle regression.
|
|
|
Portfolio Management with Multiple Benchmarks
Navrátil, Robert ; Večeř, Jan (advisor) ; Pešta, Michal (referee)
Portfolio Management with Multiple Benchmarks Bc. Robert Navrátil Abstract: In this thesis, we study a maximal volatility portfolio that treats all assets in a symmetric way and related option contract. To preserve symmetry we need numeraire that treats all assets symmetrically. We choose market index with equal weights. In case of two assets we focus on a variation of a passport option on the portfolio. The optimal strategy for the investor is the mentioned maximal volatility portfolio. We extend the known optimal strategy for the option to a richer class of convex payoff functions. We also show a modification of the optimal strategy for maximizing the probability of ending above or at a desired level. We later extend the symmetric market model to case of three assets, which can be even further extended to an arbitrary number of assets. The three asset model requires more parameters than are observable from the data, however we show indistinguishably of the model on the choice of parameters under very natural conditions. Both numerical simulations and an application on real data is provided. 1
|
|
|
Statistical inference in varying coefficient models
Splítek, Martin ; Maciak, Matúš (advisor) ; Pešta, Michal (referee)
Tato práce se zabývá modely s promìnlivými koe cienty se za- mìøením na statistickou inferenci. Hlavní my¹lenkou tìchto modelù je pou¾ití regresních koe cientù, mìnících se v závislosti na nìjakém modi kátoru vlivu, namísto konstantních koe cientù klasické lineární regrese. Nejprve si de nujeme tyto modely a jejich odhadové procedury, kterých bylo doposud publikováno nì- kolik variant. K odhadu se pou¾ívá lokální regrese nebo rùzné druhy splajnù { vyhlazovací, polynomiální èi penalizované. Od metody odhadu se následnì od- víjí i daná statistická inference, ke které uvedeme odvozené vychýlení, rozptyl, asymptotickou normalitu, kon denèní pásma a testování hypotéz. Hlavním cílem na¹í práce je kompaktnì shrnout vybrané metody a jejich inferenci. Na závìr je navr¾ena proceduru pro výbìr promìnných.
|
|
|
Tweedie models for pricing and reserving
Smolárová, Tereza ; Pešta, Michal (advisor) ; Cipra, Tomáš (referee)
This presented thesis deals with applications of Tweedie compound Poisson model in non-life insurance pricing and claims reserving. Tweedie models are exponen- tial dispersion models with power mean-variance relationships and compound Poisson distribution is a particular Tweedie model. The interest in Tweedie com- pound Poisson model is motivated by its applications to generalized linear models (GLMs) and generalized estimation equations (GEE). The purpose of this thesis is to construct pricing and claims reserving models in which the response variables follow Tweedie compound Poisson model. Theoretical approaches are applied on the real datasets. 1
|
guest ::
RSS feed.