arXiv subjects

Ronald Hochreiter

Publications and source records attributed to Ronald Hochreiter.

At least 19 recordsLinked to original sources

arXiv2015

Efficient and robust calibration of the Heston option pricing model for American options using an improved Cuckoo Search Algorithm

In this paper an improved Cuckoo Search Algorithm is developed to allow for an efficient and robust calibration of the Heston option pricing model for American options. Calibration of stochastic volatility models like the Heston is significantly harder than classical option pricing models as more parameters have to be estimated. The difficult task of calibrating one of these models to American Put options data is the main objective of this paper. Numerical results are shown to substantiate the suitability of the chosen method to tackle this problem.

cs.NE↗

arXiv2015

Computing a consensus journal meta-ranking using paired comparisons and adaptive lasso estimators

In a "publish-or-perish culture", the ranking of scientific journals plays a central role in assessing performance in the current research environment. With a wide range of existing methods and approaches to deriving journal rankings, meta-rankings have gained popularity as a means of aggregating different information sources. In this paper, we propose a method to create a consensus meta-ranking using heterogeneous journal rankings. Using a parametric model for paired comparison data we estimate quality scores for 58 journals in the OR/MS community, which together with a shrinkage procedure allows for the identification of clusters of journals with similar quality. The use of paired comparisons provides a flexible framework for deriving a consensus score while eliminating the problem of data missingness.

stat.AP↗

arXiv2015

Computing trading strategies based on financial sentiment data using evolutionary optimization

In this paper we apply evolutionary optimization techniques to compute optimal rule-based trading strategies based on financial sentiment data. The sentiment data was extracted from the social media service StockTwits to accommodate the level of bullishness or bearishness of the online trading community towards certain stocks. Numerical results for all stocks from the Dow Jones Industrial Average (DJIA) index are presented and a comparison to classical risk-return portfolio selection is provided.

q-fin.PM↗

arXiv2015

An Evolutionary Optimization Approach to Risk Parity Portfolio Selection

In this paper we present an evolutionary optimization approach to solve the risk parity portfolio selection problem. While there exist convex optimization approaches to solve this problem when long-only portfolios are considered, the optimization problem becomes non-trivial in the long-short case. To solve this problem, we propose a genetic algorithm as well as a local search heuristic. This algorithmic framework is able to compute solutions successfully. Numerical results using real-world data substantiate the practicability of the approach presented in this paper.

q-fin.PM↗

arXiv2014

The Role of Emotions in Propagating Brands in Social Networks

A key aspect of word of mouth marketing are emotions. Emotions in texts help propagating messages in conventional advertising. In word of mouth scenarios, emotions help to engage consumers and incite to propagate the message further. While the function of emotions in offline marketing in general and word of mouth marketing in particular is rather well understood, online marketing can only offer a limited view on the function of emotions. In this contribution we seek to close this gap. We therefore investigate how emotions function in social media. To do so, we collected more than 30,000 brand marketing messages from the Google+ social networking site. Using state of the art computational linguistics classifiers, we compute the sentiment of these messages. Starting out with Poisson regression-based baseline models, we seek to replicate earlier findings using this large data set. We extend upon earlier research by computing multi-level mixed effects models that compare the function of emotions across different industries. We find that while the well known notion of activating emotions propagating messages holds in general for our data as well. But there are significant differences between the observed industries.

cs.SI↗

arXiv2014

Active extension portfolio optimization with non-convex risk measures using metaheuristics

We consider the optimization of active extension portfolios. For this purpose, the optimization problem is rewritten as a stochastic programming model and solved using a clever multi-start local search heuristic, which turns out to provide stable solutions. The heuristic solutions are compared to optimization results of convex optimization solvers where applicable. Furthermore, the approach is applied to solve problems with non-convex risk measures, most notably to minimize Value-at-Risk. Numerical results using data from both the Dow Jones Industrial Average as well as the DAX 30 are shown.

q-fin.PM↗

arXiv2014

Effects of Sampling Methods on Prediction Quality. The Case of Classifying Land Cover Using Decision Trees

Clever sampling methods can be used to improve the handling of big data and increase its usefulness. The subject of this study is remote sensing, specifically airborne laser scanning point clouds representing different classes of ground cover. The aim is to derive a supervised learning model for the classification using CARTs. In order to measure the effect of different sampling methods on the classification accuracy, various experiments with varying types of sampling methods, sample sizes, and accuracy metrics have been designed. Numerical results for a subset of a large surveying project covering the lower Rhine area in Germany are shown. General conclusions regarding sampling design are drawn and presented.

stat.ML↗

arXiv2014

Modeling multi-stage decision optimization problems

Multi-stage optimization under uncertainty techniques can be used to solve long-term management problems. Although many optimization modeling language extensions as well as computational environments have been proposed, the acceptance of this technique is generally low, due to the inherent complexity of the modeling and solution process. In this paper a simplification to annotate multi-stage decision problems under uncertainty is presented - this simplification contrasts with the common approach to create an extension on top of an existing optimization modeling language. This leads to the definition of meta models, which can be instanced in various programming languages. An example using the statistical computing language R is shown.

math.OC↗

arXiv2014

Data Mining Cultural Aspects of Social Media Marketing

For marketing to function in a globalized world it must respect a diverse set of local cultures. With marketing efforts extending to social media platforms, the crossing of cultural boundaries can happen in an instant. In this paper we examine how culture influences the popularity of marketing messages in social media platforms. Text mining, automated translation and sentiment analysis contribute largely to our research. From our analysis of 400 posts on the localized Google+ pages of German car brands in Germany and the US, we conclude that posting time and emotions are important predictors for reshare counts.

cs.SI↗

arXiv2014

Automated Classification of Airborne Laser Scanning Point Clouds

Making sense of the physical world has always been at the core of mapping. Up until recently, this has always dependent on using the human eye. Using airborne lasers, it has become possible to quickly "see" more of the world in many more dimensions. The resulting enormous point clouds serve as data sources for applications far beyond the original mapping purposes ranging from flooding protection and forestry to threat mitigation. In order to process these large quantities of data, novel methods are required. In this contribution, we develop models to automatically classify ground cover and soil types. Using the logic of machine learning, we critically review the advantages of supervised and unsupervised methods. Focusing on decision trees, we improve accuracy by including beam vector components and using a genetic algorithm. We find that our approach delivers consistently high quality classifications, surpassing classical methods.

cs.CE↗

arXiv2014

Modeling Credit Spreads Using Nonlinear Regression

The term structure of credit spreads is studied with an aim to predict its future movements. A completely new approach to tackle this problem is presented, which utilizes nonlinear parametric models. The Brain-Cousens regression model with five parameters is chosen to describe the term structure of credit spreads. Further, we investigate the dependence of the parameter changes over time and the determinants of credit spreads.

q-fin.ST↗

arXiv2014

Zombie Politics: Evolutionary Algorithms to Counteract the Spread of Negative Opinions

This paper is about simulating the spread of opinions in a society and about finding ways to counteract that spread. To abstract away from potentially emotionally laden opinions, we instead simulate the spread of a zombie outbreak in a society. The virus causing this outbreak is different from traditional approaches: it not only causes a binary outcome (healthy vs infected) but rather a continuous outcome. To counteract the outbreak, a discrete number of infection-level specific treatments is available. This corresponds to acts of mild persuasion or the threats of legal action in the opinion spreading use case. This paper offers a genetic and a cultural algorithm that find the optimal mixture of treatments during the run of the simulation. They are assessed in a number of different scenarios. It is shown, that albeit far from being perfect, the cultural algorithm delivers superior performance at lower computational expense.

cs.SI↗

arXiv2014

An Evolutionary Approach towards Clustering Airborne Laser Scanning Data

In land surveying, the generation of maps was greatly simplified with the introduction of orthophotos and at a later stage with airborne LiDAR laser scanning systems. While the original purpose of LiDAR systems was to determine the altitude of ground elevations, newer full wave systems provide additional information that can be used on classifying the type of ground cover and the generation of maps. The LiDAR resulting point clouds are huge, multidimensional data sets that need to be grouped in classes of ground cover. We propose a genetic algorithm that aids in classifying these data sets and thus make them usable for map generation. A key feature are tailor-made genetic operators and fitness functions for the subject. The algorithm is compared to a traditional k-means clustering.

cs.NE↗

arXiv2014

A Genetic Algorithm to Optimize a Tweet for Retweetability

Twitter is a popular microblogging platform. When users send out messages, other users have the ability to forward these messages to their own subgraph. Most research focuses on increasing retweetability from a node's perspective. Here, we center on improving message style to increase the chance of a message being forwarded. To this end, we simulate an artificial Twitter-like network with nodes deciding deterministically on retweeting a message or not. A genetic algorithm is used to optimize message composition, so that the reach of a message is increased. When analyzing the algorithm's runtime behavior across a set of different node types, we find that the algorithm consistently succeeds in significantly improving the retweetability of a message.

cs.NE↗

arXiv2014

Evolving Accuracy: A Genetic Algorithm to Improve Election Night Forecasts

In this paper, we apply genetic algorithms to the field of electoral studies. Forecasting election results is one of the most exciting and demanding tasks in the area of market research, especially due to the fact that decisions have to be made within seconds on live television. We show that the proposed method outperforms currently applied approaches and thereby provide an argument to tighten the intersection between computer science and social science, especially political science, further. We scrutinize the performance of our algorithm's runtime behavior to evaluate its applicability in the field. Numerical results with real data from a local election in the Austrian province of Styria from 2010 substantiate the applicability of the proposed approach.

cs.NE↗

arXiv2014

An algorithm for calculating steady state probabilities of $M|E_r|c|K$ queueing systems

This paper presents a method for calculating steady state probabilities of $M|E_r|c|K$ queueing systems. The infinitesimal generator matrix is used to define all possible states in the system and their transition probabilities. While this matrix can be written down immediately for many other $M|PH|c|K$ queueing systems with phase-type service times (e.g. Coxian, Hypoexponential, \ldots), it requires a more careful analysis for systems with Erlangian service times. The constructed matrix may then be used to calculate steady state probabilities using an iterative algorithm. The resulting steady state probabilities can be used to calculate various performance measures, e.g. the average queue length. Additionally, computational issues of the implementation are discussed and an example from the field of telecommunication call-center queue length will be outlined to substantiate the applicability of these efforts. In the appendix, tables of the average queueing length given a specific number of service channels, traffic density, and system size are presented.

eess.SY↗

arXiv2014

Evolutionary Optimization for Decision Making under Uncertainty

Optimizing decision problems under uncertainty can be done using a variety of solution methods. Soft computing and heuristic approaches tend to be powerful for solving such problems. In this overview article, we survey Evolutionary Optimization techniques to solve Stochastic Programming problems - both for the single-stage and multi-stage case.

cs.NE↗

arXiv2014

Revolutionary Algorithms

The optimization of dynamic problems is both widespread and difficult. When conducting dynamic optimization, a balance between reinitialization and computational expense has to be found. There are multiple approaches to this. In parallel genetic algorithms, multiple sub-populations concurrently try to optimize a potentially dynamic problem. But as the number of sub-population increases, their efficiency decreases. Cultural algorithms provide a framework that has the potential to make optimizations more efficient. But they adapt slowly to changing environments. We thus suggest a confluence of these approaches: revolutionary algorithms. These algorithms seek to extend the evolutionary and cultural aspects of the former to approaches with a notion of the political. By modeling how belief systems are changed by means of revolution, these algorithms provide a framework to model and optimize dynamic problems in an efficient fashion.

cs.NE↗