Thin and heavy tails in stochastic programming

Kaňková, Vlasta; Houda, Michal

doi:10.14736/kyb-2015-3-0433

Cited by 8 publications

(27 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…An index which is not dominated by any other index, as well as, not by any other additional 7 It is obvious that the findings of this study depend on the sample and its related distribution functions. This relationship, as well as the empirical estimates of stochastic optimization, is examined in the papers of Kankova and Houda (2015) and Kankova and Omelchenko (2015).…”

Section: Methodsmentioning

confidence: 99%

SSD Efficiency at Multiple Data Frequencies: Application on the OECD Countries

Uğurlu¹,

Taş²,

Güran³

et al. 2018

Prague Economic Papers

View full text Add to dashboard Cite

The second order stochastic dominance (SSD) has become exceedingly popular in recent years, due to its ability to determine the dominance of one asset over another for all risk-averse investors without a strict requirement in asset distribution. In this study, 33 OECD country indexes and their enriched set of assets, which consists of some combinations of these indexes, are investigated and compared between 2007 and 2015 by utilizing pairwise SSD comparisons, with different data frequencies, such as daily, weekly, monthly and quarterly. This paper contributes to the literature in three points: Firstly, a serious portion of the best performing OECD countries has the lowest GDP (PPP) per capita level. Secondly, the SSD efficient set depends on data frequency. Thirdly, when the data frequency is lowered, the difference between two SSD pairwise efficiency tests decreases.

show abstract

Section: Methodsmentioning

confidence: 99%

SSD Efficiency at Multiple Data Frequencies: Application on the OECD Countries

Uğurlu¹,

Taş²,

Güran³

et al. 2018

Prague Economic Papers

View full text Add to dashboard Cite

show abstract

“…For the conclusion (b), employing (8), the closedness and continuity of S 1 on (P p (Ω), σ p ) are derived by that ones of S 1 on (P p (Ω), σ p ). Now, for all µ, ν ∈ P p (Ω), let x ∈ S 1 (µ) and x ∈ S 1 (ν), we have Hence, the proof is complete.…”

Section: Stochastic Optimization Problemsmentioning

confidence: 99%

“…Various settings similar to these two examples can be also modelled as the stochastic optimization problems. Therefore, the investigation of stochastic optimization has become one of the interesting and important topic in optimization theory and applications, including existence conditions [1][2][3][4][5][6], stability conditions [7][8][9] and solution methods [10][11][12]. This fact has inspired many mathematicians to study various generalized stochastic problems related to optimization with numerious applications to real-world problems, such as stochastic variational inequalities [13][14][15][16][17], stochastic Nash equilibrium problems [18][19][20][21], stochastic mathematical programs with equilibrium constraints [22][23][24] with numerious applications to real-world problems.…”

Section: Introductionmentioning

confidence: 99%

On the existence and stability of solutions to stochastic equilibrium problems

et al. 2021

View full text Add to dashboard Cite

In this paper we consider stochastic equilibrium problems involving parameter of probability measures. Employing KKM-Fan xed point theorem, conditions for the existence of solutions to such problems are established. We then propose new metric concepts on the underlying stochastic spaces and study some properties corresponding to these metrics. Afterwards, we study sucient conditions for the solution mappings of such problems, that are closed, upper (lower) semicontinuous and continuous with respect to the mentioned metrics. Finally, the special cases of stochastic optimization problems are taken into account as the applications.

show abstract

“…Proposition 2.13. (Kaňková and Houda [14]) Let X be a compact set, P F ∈ M 1 1 (R s ), Assumptions A.0, A.1, A.2 and A.3 be fulfilled, X F be defined by the relation (1.5). Let, moreover, g(x, z) be for every x ∈ X a Lipschitz function of z ∈ Z F with the Lipschitz constant not depending on x ∈ X. If…”

Section: Empirical Estimatesmentioning

confidence: 99%