Πλοήγηση ανά Επιβλέπων "Kavussanos, Manolis"

A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

Α Β Γ Δ Ε Ζ Η Θ Ι Κ Λ Μ Ν Ξ Ο Π Ρ Σ Τ Υ Φ Χ Ψ Ω

Τώρα δείχνει 1 - 9 από 9

Are co-skewness and co-kurtosis factors priced? Evidence from the UK and Greek stock markets
(14-12-2015) Stefou, Eleni; Athens University of Economics and Business, Department of Accounting and Finance; Kavussanos, Manolis
This study explores the role of higher moments on the UK and Greek stock market using the asset pricing framework developed in Fang and Lai (1997). Estimations have been made for the whole period and two sub-periods for both markets. The models estimated are: 1) Four-Moment CAPM: 𝑅𝑖=𝑏0+𝑏1𝛽𝑖+𝑏2𝛾𝑖+𝑏3𝛿𝑖 2) Three-Moment CAPM: 𝑅𝑖=𝑏0+𝑏1𝛽𝑖+𝑏2𝛾𝑖 , 3) Two-Moment CAPM: 𝑅𝑖=𝑏0+𝑏1𝛽𝑖 and 4) CAPM with beta and kurtosis: 𝑅𝑖=𝑏0+𝑏1𝛽𝑖+𝑏3𝛿𝑖. We present the estimated premiums and their t-statistics. The results show that for most portfolios the asset pricing performance improves when we add skewness and/or kurtosis to the CAPM. In addition to the grouping procedure suggested by Fang and Lai (1997) , three different grouping procedures ate tested on the UK market. In these cases while the models provide significant estimates the explanatory power of the models remains low.
Efficiency of Forward Freight Agreements: the “Unbiasedness Hypothesis"
Liokis, Dimitrios; Athens University of Economics and Business, Department of Accounting and Finance; Kavussanos, Manolis
This paper investigates the “Unbiasedness Hypothesis” of Forward Freight Agreement (FFA) testing the hypothesis for dry-bulk FFAs of individual routes P2A, P3A, C4 and C7 one, two and three month prior to their maturity. Capesize 4TC, Panamax 4TC and Supramax 6TC are also investigated. A VECM is estimated by using the framework introduced by Johanssen (1988) to test for cointegration. Unbiasedness is tested by putting the necessary restrictions in the cointegrating vector and using the statistic introduced by Juselius and Johansen (1990). Unbiased FFAs are a free source of information about the expected future spot prices, being beneficial to any economic agent exposed to the freight market.
Essays on finance, economics and shipping
(06/29/2021) Moysiadou, Stergiani A.; Μωϋσιάδου, Στεργιανή; Athens University of Economics and Business, Department of Accounting and Finance; Drakos, Konstantinos; Rompolis, Leonidas; Tsekrekos, Andrianos; Georgoutsos, Dimitrios; Chalamandaris, George; Tsouknidis, Dimitrios; Kavussanos, Manolis
This Ph.D. Thesis consists of four research papers. The first paper provides a brief and concise description of the various sectors and segments of the shipping industry and the main economic participants. It also analyses the factors affecting supply and demand in the dry and wet bulk sectors of the industry, and how these forces interact to determine freight rate equilibrium both in the short and the long run, in the various shipping markets (i.e. the freight, the newbuilding, the sales and purchase and the demolition markets). The reader can learn about key shipping market concepts following the historical evolution of shipping cycles in response to core events that triggered changes in market dynamics over time.The second paper examines the long-run market efficiency hypothesis in the bulk shipping transportation markets, observing long-run equilibrium relationships between freight rates on different shipping routes and the dynamics of short-run deviations from the equilibria. The freight rate series on the different shipping routes examined are selected on the basis of a common economic driver, which may be a common port of import, export or a common commodity transported. The results of the analysis confirm the long-run market efficiency hypothesis, providing useful insights for the efficient organisation of supply-chain models. The third paper studies the effect that the relative market power between buyers and sellers of the shipping transportation service (that is, vessel charterers and owners, respectively) may have on the formation on individual voyage freight rates. Also, the paper highlights the asymmetry of the market power distribution between charterers and shipowners, which is essential to consider especially in cases of prospective mergers or acquisitions between chartering companies with high market shares, due the potential adverse consequences such mergers may have on the relevant freight markets. The fourth paper calibrates and examines the performance of various Value-at-Risk (VaR) and Expected Shortfall (ES) modelling specifications for the prediction of ETF risk in dry bulk shipping. The study focuses on ETFs priced against and investing in shipping freight derivatives in the dry bulk sector of the industry. The paper highlights the importance of the accurate estimation of the risk that investing in newly-introduced financial products (like the one under investigation) entails, products which provide access to the freight markets for both individual and institutional investors.
The impact of freight rates on the stock prices of US listed shipping companies
(2020) Diamantis, Nektarios; Διαμαντής, Νεκτάριος; Karavasilis, Nikolaos; Καραβασίλης, Νικόλαος; Demirakos, Efthimios; Rompolis, Leonidas; Kavussanos, Manolis
The purpose of this paper is to investigate and present the impact of freight rates on the stock prices of shipping companies. In our research, we are focusing on this particular impact on US listed shipping companies during the five-year period between January 2015 and December 2019. We are going to examine whether there is any kind of relation between stock prices returns and shipping freight rates in order to provide further information to the existing bibliography of shipping stock prices forecasting. By taking into consideration previous literature, we are observing that there is a strong relation between these two, although it is highlighted the fact that besides freight rates there are a lot of other factors enhancing the volatility that shipping stocks are experiencing. We are going to take the existing research a step further by estimating the level of repercussion of freight rates on the stock prices and also we are going to consider other important variables that due to their substance contain the ability to explain in a significant proportion the latter. In order to implement the scope of the aforementioned discussion we use various databases to collect the relevant data. We conduct our analysis with the assistance of statistical tools from where we derive our results. The main findings we extracted from the previous procedure agree with the existing literature and more specifically result in an important and meaningful interconnection inside our data set.
Measuring market risk in financial and freight markets
Dimitrakopoulos, Dimitris; Athens University of Economics and Business, Department of Accounting and Finance; Kavussanos, Manolis
The recent global financial crisis has amply highlighted the importance of prudent management of financial risks for the well functioning of firms and the prevention of a systemic crisis. Risk measurement constitutes an mandatory step towards risk management which is linked tightly to other layers of the risk management process of such as risk reporting, limit setting, performance evaluation and risk budgeting. This thesis focuses on market risk by investigating the critical issue of market risk measurement. A main concern for risk taking agents and regulators is the accurate estimation of market risk which is directly linked to the process of the selection of a risk measurement model. The latter turns out to be a matter of empirical investigation depending on several factors complicating decisions on the specification of a preferable risk model, such as the properties the underlying assets for which market risk is estimated, the time period and the investment horizon over which the risk measurement models are estimated and the confidence level of risk estimations, amongst others. Additionally, although financial literature on market risk measurement has progressed substantially during the last decade leading to the development of a vast number of market risk measurement models and validation techniques, it remains inconclusive as to which method is preferable for the estimation of market risk. The aim of this thesis is to investigate the critical issue of market risk quantification by shedding some light on the controversial issue of market risk model selection. For this purpose two alternative risk metrics were utilized: The value at risk (VaR) and the expected tail loss (ETL). Broadly speaking, VaR and ETL are summary statistics that quantify in a probabilistic manner the exposure of an asset or portfolio to market risks. More specifically, VaR measures the loss that shall not be exceeded with a given probability over a fixed investment horizon; see for e.g. Jorion (1997). The ETL is defined as the expectation of the loss beyond the VaR level; see for e.g. Artzner et al. (1997). The markets chosen for the empirical analysis have a prominent role in the world economy and are extremely important for the global economic stability. Furthermore, little or no empirical research exists investigating the issue of market risk measurement for these markets. These are emerging equity markets and ocean freight markets for the liquid and the dry bulk segments of shipping. In order to elucidate different patterns in the behavior of the market risk measurement models and revisit the documented performance of the alternative methods in the case of mainstream financial assets, an additional sample of four developed markets equity indices was included. To this end, the employed sample consists of 18 international financial corporation emerging stock market indices as well as 4 developed equity market indices from the regions of Asia, America and Europe adding up to a diverse and representative sample of emerging and developed equity portfolios. As regards the dry bulk and the liquid bulk freight rate markets, aggregate and specific route indices of the Baltic Exchange are considered. A large number of VaR and ETL forecasting models are compared in an empirical evaluation assessment. These include: traditional models such as the random walk, the riskmetrics, GARCH variants and historical simulation based methods, extreme value methods such as the peaks over threshold and the block maxima method, and hybrid methods such as the EGARCH-VaR-X approach. These models were evaluated in terms of statistical accuracy by employing Christoffersen’s 1998, Lopez (1998) and Harvey et al. (1997) backtesting methods. There are certain idiosyncratic and in some cases unique attributes that pertain to the markets considered rendering the investigation of market risk measurement particularly interesting and challenging. These involve the turbulent nature of emerging and freight markets, the low correlation of emerging markets with the developed markets, the contagion effect in between emerging markets during periods of financial turmoil, the highly cyclical and seasonal behavior of freight rate returns and the longer term investment horizons being part of risk taking agents concerns who activate in freight rate markets. This thesis makes a number of important contributions: first, it demonstrates which market risk measurement methods are appropriate for the aforementioned markets. The importance of these markets for the world economy in conjunction with the fact that little or no empirical research exists investigating the issue of market risk measurement renders the investigation of this thesis particularly interesting; second, it addresses the issues of market risk estimation in association with the unique attributes present in these markets; third, it revisits the issue of market risk model selection for developed equity markets, allowing direct comparisons to be made with the results obtained from the assets investigated in this thesis; fourth, additional issues related to the risk measurement such as the estimation of medium term market risk and model selection during periods of crises or when the data generating process incorporates seasonality and cyclicality are addressed. Fifth, findings may guide decisions regarding risk estimation for other markets which share common characteristics with the aforementioned markets, such as the electricity and the agricultural commodity markets. The findings of this thesis can be summarized in the following: Overall, it was found that the fatter the tail of the empirical profit/loss distribution, the more difficult the estimation of market risk becomes. Specifically, few or no models managed to survive the backtesting for the markets which exhibit the fattest tails as measured by the tail index. This highlights the importance of using extreme value based diagnostic tools (such as the tail index estimates used in this thesis) in the practice of risk management to determine the properties of the tails of the profit/loss distributions and adapt risk modeling accordingly. The most successful selected VaR models were common for both emerging and developed equity markets despite the documented differences between these two asset classes standing out. However, in contrast to many studies on emerging markets selecting extreme value market risk measurement methods, the hereby thesis selects the simple non-parametric historical simulation based method of Boudoukh et al. (1998). In general, it was found that the tradeoff between statistical accuracy and complexity in the estimation of VaR was advantageous in the case of non-parametric models; in other words, these models seem to provide statistically sufficient risk forecasts just like the extreme value or other sophisticated VaR models do with comparatively fewer resources needed for their estimation. However, it is notable that although the same VaR model was selected for both emerging and developed markets, diversities concerning the performance of the same VaR models with respect to the estimation window were demonstrated: Specifically, most VaR models tend to overestimate and underestimate the realized VaR for the emerging and developed markets, respectively. Thus, if long estimation windows are employed for the estimation of the market risk in an emerging markets context, risk reporting for imposing capital reserve requirements may turn out to be costly due to the overestimation of risk. Another subtle point which deserves some attention is that special care should be taken in the case of post crisis periods during which the performance of the VaR models which included the crisis in the estimation sample deteriorated substantially. Assuming that this finding describes a typical pattern in the estimation of VaR during post-crisis periods, then it is very significant for the periods succeeding the current financial crisis as it suggests that most VaR models incorporating the crisis in their estimation sample, are likely to produce inaccurate risk forecasts. In the case of ocean freight rates findings indicate strong segregation between the markets examined as the performance of the alternative risk measurement models differed significantly leading to the selection of different models for each market. This finding contradicts literature for the majority of asset classes where the alternative risk measurement methods seem to perform similarly across the various assets belonging to the same asset class (i.e. equity portfolios of emerging and developed markets). In general, despite the salient unique empirical regularities evident in freight rate markets, it was found that the liquid bulk freight rates can be properly modeled as conventional financial assets for risk measurement purposes. However, different risk measurement models are selected for quantifying market risk in the liquid bulk freight markets: Specifically, the best performing models for quantifying daily liquid bulk freight rate risk exposures were the historical simulation, the GARCH, and the random walk VaR and ETL methods. In contrast to the liquid bulk freight rates, in the case of risk quantification for the dry bulk freight rates most conventional VaR and ETL models fell short in providing sufficiently accurate risk forecasts, demanding different treatment as far as market risk estimation is concerned. To this end, more sophisticated risk measurement models such as the switching ARCH, the filtered peaks over threshold and the combination of an exponential GARCH with student’s-t innovations parameterized by the Huisman et al. (2001) algorithm, outperformed other models in terms of statistical sufficiency and accuracy. Finally, this thesis makes an attempt to resolve the problem of estimating longer term risk with limited data by comparing the scaling laws of the square root and the tail index root of time with an empirically determined scaling law based on the theory of self similarity. The latter is proven to yield the most accurate medium term risk forecasts in contrast to scaling by the square or the tail index root of time which were found to understate the realized VaR systematically. As medium term risk exposures are relevant in many applications in finance (for e.g. when estimating market risk for imposing capital reserve requirements or in the case of estimating risk of insurance company portfolios) the proposed scaling law can be used effectively for extrapolating longer term risk forecasts with high frequency data by increasing the accuracy of risk forecasts.
Multiple factor analyses of the perceived market risk in the maritime industry a decade after 2008 global financial
(11/30/2018) Katsiada, Aikaterini; Kokorogianni, Artemis; Athens University of Economics and Business, Department of Accounting and Finance; Chalamandaris, George; Demirakos, Efthymios; Kavussanos, Manolis
The aim of this master thesis is to develop multiple models in order to investigate the determining factors that dominated the shipping industry’s market risk a decade after global financial crisis of 2008. To this end, this analysis uses a sample of shipping firms listed in US, their betas and other financial characteristics, as well as the outstanding macroeconomic characteristics of the global market over the period January 2009 – October 2018. Yet, it is attempted to compare and contrast the dominant factors over the first five years after the 2008 crisis commenced (up to 2013), versus the latter five years (from 2013 on). The set of the microeconomic consists of: market value-to-book value of equity ratio; current ratio; leverage as described by debt-to-asset ratio; dividend yield; earnings growth; ROE; and ROI; while the macroeconomic factors used are: bunker prices; inflation and industrial production for OECD countries and world steel production. The power of the relationships described above are investigated through panel regression analysis for the firm specific variables, while the macroeconomic variables, which are time-series, are tested through time series regression analysis. The highlight points in our findings are that both micro and macro-economic factors seem to play a significant role affecting the shipping firms’ betas, although the most important factors for the first period (2009-2013) are industrial production growth, financial leverage and return on equity ratio (ROE). Nevertheless, for the second period (2014-2018) the bunker prices, current ratio and financial leverage seem to be more powerful in determining a shipping firm’s market risk.
Predicting spot freight rates by spreads of time charter contracts of different duration
(2020) Tsiouprou, Maria; Τσιούπρου, Μαρία; Athens University of Economics and Business, Department of Accounting and Finance; Moutos, Thomas; Demirakos, Efthimios; Kavussanos, Manolis
This dissertation examines the predictive ability of spreads between time charter contracts of different durations to spot freight rates. As the spot freight rates are characterized by extreme volatility, the need to predict them is undeniable. In this research monthly data were collected from Clarksons Shipping Intelligence Network for a period starting from January 2009 until September 2020. The analysis refers to three different ship types, which are Capesize, Panamax and Handysize vessels. For each vessel type, a linear model was defined from a pool of variables that, based on theory, affect freight rates. Then, the first model was compared with the same model with a time charter spread added.
Risk measurement in the capesize market
(02/23/2021) Drakopoulou, Argyro; Δρακοπούλου, Αργυρώ; Mavrikou, Georgia; Μαυρίκου, Γεωργία; Athens University of Economics and Business, Department of Accounting and Finance; Moutos, Thomas; Demirakos, Efthimios; Kavussanos, Manolis
The present paper aims to study the management of risk in the Capesize market. It instigates this aim via the study of the secondary sources through a literature review, specifically of the field of the Capesize market and the shipping cycles, as well as the concepts of risk management and risk analysis. Following that, the paper expands on the methodology it followed, presents its analysis, and results and finally presents its conclusions.
Valuation of a shipping company
Σκόνδρας, Κωνσταντίνος; Athens University of Economics and Business, Department of Accounting and Finance; Kavussanos, Manolis
Valuation of companies has always been a sector, where many people in economics devoted their concerns, about finding the best way to value a company.For that reason, many different ways have been proposed to find the “fair” value of a company. However, the process of valuation is a demanding task and sometimes,dependent on different theoretical frameworks and subject to the analysts’ subjective interpretations. This dissertation is focused on valuing two listed shipping companies of a different sector. The first company is Diana Containerships Inc. a Greek shipping company and the second company is Capital Product Partners L.P, both listed on Nasdaq, Thus, providing an illustration of the challenges faced when valuing enterprises. In particular, three different theoretical frameworks, suggested by the most renowned authors in this field of study, were utilized to capture the fair values of the companies. To successfully fulfill the purpose of this thesis, state of the art literature was reviewed, in order to present the valuation models.