Conference Agenda

Economists have conducted research on several empirical phenomena regarding the behavior of individual investors, such as how their emotions and opinions influence their decisions. All those emotions and opinions are described by the word Sentiment. In finance, stochastic changes might occur according to investors sentiment levels. In this study, our main goal is to apply several operational research techniques and analyze these techniques’ accuracy. Firstly, we represent the mutual effects between some financial process and investors sentiment with multivariate adaptive regression splines (MARS) model. Furthermore, we consider to extend this model by using distinct data mining techniques and compare the gain in accuracy and computational time with its strong alternatives applied in the analyses of the financial data. Hence, the goal of this study is to compare the forecasting performance of sentiment index by using two-stage MARS-NN (neural network), MARS-RF (random forest), RF-MARS, RF-NN, NN-MARS, and NN-RF hybrid models. Furthermore, we aim to classify the peoples’ feelings about economy according to their confidence levels. Moreover, to forecast the underlying state change of the consumer confidence index (CCI) and to observe the relationship with some macroeconomic data (CPI, GDP and currency rate) at a monthly interval, we apply hidden Markov model (HMM). The aim is to detect the switch between these states and to define a path of these states. We also aim to use volatility models for mainly sentiment index, consumer confidence index, and other indices so that we can get better forecasting results from those datasets.

This study examines the diversification benefits and hedging abilities of cryptocurrencies like Bitcoin for German stock market investors. From an economic perspective, cryptocurrencies are interpreted as an additional asset class and incorporated into portfolios of equities. Using German stock market data and Bitcoin data from January 2012 to December 2022, the yearly addition of Bitcoin to different long equities portfolios is analysed. To ensure the independence of the results from assumptions regarding investors' risk preferences, portfolios of equities and Bitcoin were constructed with the same projected standard deviation of return as for the pure stock portfolio. Performance measures, such as the reward-to-variability ratio, and risk measures, such as the one-sided Value-at-Risk, were applied to the portfolios created. Furthermore, Bitcoin was employed as a means of hedging equities, with the objective of mitigating the impact of positive and negative stock market movements. In this section, multiple regressions are estimated. The findings indicate that portfolios comprising equities and Bitcoin exhibit enhanced performance relative to portfolios comprising equities alone. However, the limited efficacy of Bitcoin as a means of absorbing market downturns suggests that it may not be an optimal solution for absorbing market volatility. Consequently, while Bitcoin can provide substantially enhanced portfolio risk-return exposures, it is constrained in its ability to act as an efficient hedge to the stock market. Subsequently, we examine the safe haven properties of Bitcoin.

Quantum computing (QC) is attracting increasing attention as a potential portfolio optimization tool for institutional investors. To understand the implications of QC for investors' portfolio allocation, it is to recognise that QC heralds a fundamental shift in the specification of the problem. Traditionally, portfolio optimization problems have been approached predominantly with continuous specifications, due to the accessibility of efficient solvers even in high-dimensional spaces. However, discrete formulations have attracted attention due to their inherent advantages, including the ability to address cardinality constraints, logical constraints, and incorporate transaction costs. The potential benefits of discrete specification have been hampered by computational difficulties, as these problems are typically NP-hard to solve. With QC operating in binary variables, which can be easily extended to discrete problems, and its potential scalability to large dimensions, QC emerges as a promising bridge between these two approaches to portfolio optimization. This paper explores the differences between classical continuous optimisation methods and QC. It presents the first academic application of QC to a multi-objective investment problem involving preferences for sustainable investments and regulatory capital requirements for insurance companies. Using a comprehensive empirical dataset covering stocks from Europe, the US and Asia, our results suggest a promising potential of QC for institutional investors managing multiple objectives and transaction costs.

This research investigates the factors used in calculating credit scores for assessing educational and consumer microloans offered to students and recent graduates within two years of completion. The study aims to develop a credit scoring model by gathering various factors employed in credit score evaluations from existing research and surveys. Subsequently, Factor Analysis is utilized to select appropriate factors and compute the weights of the lending criteria. These factors are then incorporated into a credit scoring model, with the non-performing loan (NPL) ratio and default rate serving as efficiency indices for model evaluation. Additionally, the research simulates loan data, demonstrating the calculation of the index and offering insights on developing and improving the credit scoring model in cases with high NPL ratios and default rates.