Refine
Document Type
- Article (9)
- Doctoral Thesis (1)
Language
- English (10)
Has Fulltext
- yes (10)
Is part of the Bibliography
- no (10)
Keywords
- Cancer genomics (1)
- Di-hadron correlations (1)
- Interference fragmentation function (1)
- Long non-coding RNAs (1)
- Proteomics (1)
- Transcriptomics (1)
- Transversity (1)
- charge transfer (1)
- computational chemistry (1)
- fluorescence (1)
Institute
The transversity distribution, which describes transversely polarized quarks in transversely polarized nucleons, is a fundamental component of the spin structure of the nucleon, and is only loosely constrained by global fits to existing semi-inclusive deep inelastic scattering (SIDIS) data. In transversely polarized p↑+p collisions it can be accessed using transverse polarization dependent fragmentation functions which give rise to azimuthal correlations between the polarization of the struck parton and the final state scalar mesons.This letter reports on spin dependent di-hadron correlations measured by the STAR experiment. The new dataset corresponds to 25 pb−1 integrated luminosity of p↑+p collisions at s=500 GeV, an increase of more than a factor of ten compared to our previous measurement at s=200 GeV. Non-zero asymmetries sensitive to transversity are observed at a Q2 of several hundred GeV and are found to be consistent with the former measurement and a model calculation. We expect that these data will enable an extraction of transversity with comparable precision to current SIDIS datasets but at much higher momentum transfers where subleading effects are suppressed.
We present three-particle mixed-harmonic correlations 〈cos(mφa + nφb − (m + n)φc )〉 for harmonics m, n = 1 − 3 for charged particles in √sN N = 200 GeV Au+Au collisions at RHIC. These measurements provide information on the three-dimensional structure of the initial collision zone and are important for constraining models of a subsequent low-viscosity quark–gluon plasma expansion phase. We investigate correlations between the first, second and third harmonics predicted as a consequence of fluctuations in the initial state. The dependence of the correlations on the pseudorapidity separation between particles show hints of a breaking of longitudinal invariance. We compare our results to a number of state-of-the art hydrodynamic calculations with different initial states and temperature dependent viscosities. These measurements provide important steps towards constraining the temperature dependent viscosity and longitudinal structure of the initial state at RHIC.
Fluctuations of conserved quantities such as baryon number, charge, and strangeness are sensitive to the correlation length of the hot and dense matter created in relativistic heavy-ion collisions and can be used to search for the QCD critical point. We report the first measurements of the moments of net-kaon multiplicity distributions in Au+Au collisions at √sNN = 7.7, 11.5, 14.5, 19.6, 27, 39, 62.4, and 200 GeV. The collision centrality and energy dependence of the mean (M), variance (σ 2), skewness (S), and kurtosis (κ) for net-kaon multiplicity distributions as well as the ratio σ 2/M and the products Sσ and κσ 2 are presented. Comparisons are made with Poisson and negative binomial baseline calculations as well as with UrQMD, a transport model (UrQMD) that does not include effects from the QCD critical point. Within current uncertainties, the net-kaon cumulant ratios appear to be monotonic as a function of collision energy.
Rapidity-odd directed flow measurements at midrapidity are presented for Λ, Λ¯, K±, K0s and ϕ at sNN−−−−√= 7.7, 11.5, 14.5, 19.6, 27, 39, 62.4 and 200 GeV in Au+Au collisions recorded by the STAR detector at the Relativistic Heavy Ion Collider. These measurements greatly expand the scope of data available to constrain models with differing prescriptions for the equation of state of quantum chromodynamics. Results show good sensitivity for testing a picture where flow is assumed to be imposed before hadron formation and the observed particles are assumed to form via coalescence of constituent quarks. The pattern of departure from a coalescence-inspired sum-rule can be a valuable new tool for probing the collision dynamics.
The inclusive J/ψ transverse momentum spectra and nuclear modification factors are reported at midrapidity (|y| < 1.0) in Au+Au collisions at √sN N = 39, 62.4 and 200 GeV taken by the STAR experiment. A suppression of J/ψ production, with respect to the production in p + p scaled by the number of binary nucleon–nucleon collisions, is observed in central Au+Au collisions at these three energies. No significant energy dependence of nuclear modification factors is found within uncertainties. The measured nuclear modification factors can be described by model calculations that take into account both suppression of direct J/ψ production due to the color screening effect and J/ψ regeneration from recombination of uncorrelated charm–anticharm quark pairs.
We report the direct virtual photon invariant yields in the transverse momentum ranges 1 < pT < 3 GeV/c and 5 < pT < 10 GeV/c at mid-rapidity derived from the dielectron invariant mass continuum region 0.10 < Mee < 0.28 GeV/c2 for 0–80% minimum-bias Au+Au collisions at √sN N = 200 GeV. A clear excess in the invariant yield compared to the nuclear overlap function T A A scaled p + p reference is observed in the pT range 1 < pT < 3 GeV/c. For pT > 6 GeV/c the production follows T A A scaling. Model calculations with contributions from thermal radiation and initial hard parton scattering are consistent ithin uncertainties with the direct virtual photon invariant yield.
Glioblastoma is the most common malignant primary brain tumor. To date, clinically relevant biomarkers are restricted to isocitrate dehydrogenase (IDH) gene 1 or 2 mutations and O6-methylguanine DNA methyltransferase (MGMT) promoter methylation. Long non-coding RNAs (lncRNAs) have been shown to contribute to glioblastoma pathogenesis and could potentially serve as novel biomarkers. The clinical significance of HOXA Transcript Antisense RNA, Myeloid-Specific 1 (HOTAIRM1) was determined by analyzing HOTAIRM1 in multiple glioblastoma gene expression data sets for associations with prognosis, as well as, IDH mutation and MGMT promoter methylation status. Finally, the role of HOTAIRM1 in glioblastoma biology and radiotherapy resistance was characterized in vitro and in vivo. We identified HOTAIRM1 as a candidate lncRNA whose up-regulation is significantly associated with shorter survival of glioblastoma patients, independent from IDH mutation and MGMT promoter methylation. Glioblastoma cell line models uniformly showed reduced cell viability, decreased invasive growth and diminished colony formation capacity upon HOTAIRM1 down-regulation. Integrated proteogenomic analyses revealed impaired mitochondrial function and determination of reactive oxygen species (ROS) levels confirmed increased ROS levels upon HOTAIRM1 knock-down. HOTAIRM1 knock-down decreased expression of transglutaminase 2 (TGM2), a candidate protein implicated in mitochondrial function, and knock-down of TGM2 mimicked the phenotype of HOTAIRM1 down-regulation in glioblastoma cells. Moreover, HOTAIRM1 modulates radiosensitivity of glioblastoma cells both in vitro and in vivo. Our data support a role for HOTAIRM1 as a driver of biological aggressiveness, radioresistance and poor outcome in glioblastoma. Targeting HOTAIRM1 may be a promising new therapeutic approach.
Monte Carlo methods : barrier option pricing with stable Greeks and multilevel Monte Carlo learning
(2021)
For discretely observed barrier options, there exists no closed solution under the Black-Scholes model. Thus, it is often helpful to use Monte Carlo simulations, which are easily adapted to these models. However, as presented above, the discontinuous payoff may lead to instability in option's sensitivities for Monte Carlo algorithms.
This thesis presents a new Monte Carlo algorithm that can calculate the pathwise sensitivities for discretely monitored barrier options. The idea is based on Glasserman and Staum's one-step survival strategy and the results of Alm et al., with which we can stably determine the option's sensitivities such as Delta and Vega by finite-differences. The basic idea of Glasserman and Staum is to use a truncated normal distribution, which excludes the values above the barrier (e.g.\ for knock-up-out options), instead of sampling from the full normal distribution. This approach avoids the discontinuity generated by any Monte Carlo path crossing the barrier and yields a Lipschitz-continuous payoff function.
The new part will be to develop an extended algorithm that estimates the sensitivities directly, without simulation at multiple parameter values as in finite-difference.
Consider the local volatility model, which is a generalisation of the Black-Scholes model. Although standard Monte Carlo algorithms work well for the pricing of continuously monitored barrier options within this model, they often do not behave stably with respect to numerical differentiation.
To bypass this problem, one would generally either resort to regularised differentiation schemes or derive an algorithm for precise differentiation. Unfortunately, while the widespread solution of using a Brownian bridge approach leads to accurate first derivatives, they are not Lipschitz-continuous. This leads to instability with respect to numerical differentiation for second-order Greeks.
To alleviate this problem - i.e. produce Lipschitz-continuous first-order derivatives - and reduce variance, we generalise the idea of one-step survival to general scalar stochastic differential equations. This approach leads to the new one-step survival Brownian bridge approximation, which allows for stable second-order Greeks calculations.
To show the new approach's numerical efficiency, we present a new respective Monte Carlo pathwise sensitivity estimator for the first-order Greeks and study different methods to compute second-order Greeks stably. Finally, we develop a one-step survival Brownian bridge multilevel Monte Carlo algorithm to reduce the computational cost in practice.
This thesis proves unbiasedness and variance reduction of our new, one-step survival version with respect to the classical, Brownian bridge approach. Furthermore, we will present a new convergence result for the Brownian bridge approach using the Milstein scheme under certain conditions. Overall, these properties imply convergence of the new one-step survival Brownian bridge approach.
In recent years, deep learning has become pervasive in various fields. As a family of machine learning methods it is used in a broad set of applications, such as image processing, voice recognition, email filtering, computer vision. Most modern deep learning algorithms are based on artificial neural networks inspired by the biological neural networks constituting animal brains. Also in computational finance deep learning may be of use: Consider there is no closed-solution available for an option price, Monte Carlo simulations are substantially for estimation. Instead of persistently contributing new price computations arising from an updated volatility term, one could replace these by evaluating a neural network.
If an according neural network is available, the evaluation could lead to substantial savings and be highly efficient. I.e., once trained, a neural network could save further expensive estimations. However, in practice, the challenge is the training process of the neural network.
We study and compare two generic neural network training algorithms' computational complexity. Then, we introduce a new multilevel training algorithm that combines a deep learning algorithm with the idea of multilevel Monte Carlo path simulation. The idea is to train several neural networks with training data computed from the so-called level estimators of the multilevel Monte Carlo approach introduced by Giles. We show that the new method can reduce computational complexity by formulating a complexity theorem.
We developed three bathochromic, green-light activatable, photolabile protecting groups based on a nitrodibenzofuran (NDBF) core with D-π-A push–pull structures. Variation of donor substituents (D) at the favored ring position enabled us to observe their impact on the photolysis quantum yields. Comparing our new azetidinyl-NDBF (Az-NDBF) photolabile protecting group with our earlier published DMA-NDBF, we obtained insight into its excitation-specific photochemistry. While the “two-photon-only” cage DMA-NDBF was inert against one-photon excitation (1PE) in the visible spectral range, we were able to efficiently release glutamic acid from azetidinyl-NDBF with irradiation at 420 and 530 nm. Thus, a minimal change (a cyclization adding only one carbon atom) resulted in a drastically changed photochemical behavior, which enables photolysis in the green part of the spectrum.
The KASCADE-Grande experiment has significantly contributed to the current knowledge about the energy spectrum and composition of cosmic rays for energies between the knee and the ankle. Meanwhile, post-LHC versions of the hadronic interaction models are available and used to interpret the entire data set of KASCADE-Grande. In addition, a new, combined analysis of both arrays, KASCADE and Grande, was developed significantly increasing the accuracy of the shower observables. First results of the new analysis with the entire data set of the KASCADE-Grande experiment will be the focus of this contribution.