Search

Financial markets were the first to react on COVID-19 pandemic and were also very sensitive on following lock-outs, support policies and restrictions. Though equally exposed to the threat and lack of information, different countries reacted to the disease in a different way. In this paper we focus on European financial markets and respective representative indices to analyze their characteristics and study their response to the economic changes that followed the outbreak of the pandemic. We use six European stock market indices (FTSE 100, CAC 40, DAX, IBEX 35, FTSE MIB and STOXX Europe 600) to compare their performance before and after 2020, assess their reaction on COVID-19 and discuss their recovery. Using MLE estimation and BFGS optimization we analyze return distribution and use them as proxy for pandemic impact. [ABSTRACT FROM AUTHOR]

Recent economic, political and behavioral changes have put both traditional and modern management methods under stress. When facing supply shortages, high volatility and uncertainty about near future, it becomes crucial to be able to decide and act quickly. Unfortunately, the information that we need to decide is rarely available and requires substantial efforts and time to be collected and preprocessed. In this paper we discuss an automated system that can track and verify different information sources in order to extract valuable details and support the decision-making process. To demonstrate the benefits of such approach we analyze it the context of migration tracking in terms of crises – such like recent COVID pandemic. This process impacts not only the disease development, but it also influences industrial managers by depriving them of resources and forcing them to plan for peaks in number of sick leaves, unexpected resource shortages and delivery problems. [ABSTRACT FROM AUTHOR]

Traffic control is crucial as modern cities experience regular jams, air pollution peaks and waste of time. In this paper we present a flexible and fast approach toward optimizing city traffic with the support of available monitoring cameras, intelligent video processing algorithm and analysis of the vehicles passing by. We argue that by estimating economic impact of different optimization procedures and introducing real-time metrics on different points in the transportation network it is possible to perform a large-scale optimization and improvement of traffic with the use of available technology and minimal investments. We also suggest and test a simple technology for handling rotating cameras and automatically adjusting the tracking algorithms with regard to changes in the angle of view. [ABSTRACT FROM AUTHOR]

To create and maintain a well-functioning transportation system is a very complex task, that requires knowledge from different domains. As the number of different means of transportation grows new dependencies and dynamic behavior patterns emerge. In this paper we analyze different options for modelling transportation systems with the use of heterogeneous data inputs and sources. Our study builds on use of machine learning and big data analysis for extracting information on characteristics of transportation systems and reducing the number of incidents. We aim at developing a flexible and extensible approach, that can be adapted to transportation problems of different complexity. [ABSTRACT FROM AUTHOR]

Modelling public transportation delays is crucial for building efficient and modern city infrastructure. It is directly related to passenger behavior and satisfaction, as well as to pollution, traffic jams and sustainable development. In this paper we focus on use of jump diffusion processes in analyzing delays. We believe that such an approach is better suited for handling all factors that affect transportation and can also be used to successfully model not only the absolute time spans of waiting times, but also disproportionally large economic impact of huge delays. Our model is tested with real world data and application of obtained results is discussed. [ABSTRACT FROM AUTHOR]

Modelling and forecasting bond yields is a very important aspect in investment planning and making sound choices on different financing sources. In this paper we use time series of corporate and government bond yields in order to separate and study the impact of short- and long-term factors. Based on Nelson-Siegel ([1]) and Svensson ([2]) models, we analyse the dynamics of parameter estimates and use the results to explain market participants expectations. Unlike other studies, that aim at providing better estimates of the yield curve at specific moment in time, our major goal is to draw conclusions on the changes in the model estimates over time, as a proxy for different factors influencing bond yields. [ABSTRACT FROM AUTHOR]

Real estate is more than just an investment opportunity as it has direct impact on development of cities and places we inhabit. Being able to analyze and understand the market dynamics can help in shaping a sustainable growth and development of our cities. In this paper we present and access machine learning methods that can be used to analyze real estate price dynamics, market changes and relate them to customer preferences. We use open data to evaluate different approaches and machine (ML) and deep (DL) learning algorithms to stress on their advantages and methods to overcome their drawbacks. [ABSTRACT FROM AUTHOR]

Real estate has always been an important investment opportunity. With a diverse set of financial instruments linked to real estate assets, it is significant for both investors and intermediaries. In this paper we assess how artificial intelligence can be used to improve our understanding for the real estate market changes. We suggest and test a three-stage model in support for real estate valuation and market forecasting, that is able to account for global economic factors as well as for individual characteristics influencing property prices. Every stage provides for using different artificial intelligence and machine learning methods in order to automate processing of market data and assess how qualitative factors affect valuation. We conduct a survey on the accuracy of the model NAREIT and BGREIT index data. [ABSTRACT FROM AUTHOR]

Decision making is a complex process that requires taking into consideration multiple heterogeneous sources of uncertainty. Standard valuation and financial analysis techniques often fail to properly account for all these sources of risk as well as for all sources of additional flexibility. In this paper we explore applications of a modified binomial tree method for real options analysis (ROA) in an effort to improve decision making process. Usual cases of use of real options are analyzed with elaborate study on the applications and advantages that company management can derive from their application. A numeric results based on extending simple binomial tree approach for multiple sources of uncertainty are provided to demonstrate the improvement effects on management decisions. [ABSTRACT FROM AUTHOR]

In this paper we explore a straightforward procedure to price derivatives by using the Monte Carlo approach when the underlying process is a jump-diffusion. We have compared the Black-Scholes model with one of its extensions that is the Merton model. The latter model is better in capturing the market's phenomena and is comparative to stochastic volatility models in terms of pricing accuracy. We have presented simulations of asset paths and pricing of barrier options for both Geometric Brownian motion and exponential Levy processes as it is the concrete case of the Merton model. A desired level of accuracy is obtained with simple computer operations in MATLAB for efficient computational time. [ABSTRACT FROM AUTHOR]