Stelios Marcoulis – författare

The type of research methodologies used in analysing stock returns in the book is outlined in this technical chapter. The chapter begins with definitions of useful summary statistics, such as the mean, standard deviation, coefficient of variation, and their interpretation. Tests statistics for inferences on population means, proportions and differences of means, among others, are also presented. A summary of single equation regression techniques and the way these are applied through estimation and inference are outlined, focusing on the interpretation of standard output and diagnostic tests. The single equation methodology is then extended to multiple equation systems; the methods of Multivariate Least Square (MLSQ) of Seemingly Unrelated Regression Equations (SURE) are Regressions as well as those considered. Time series, ARlMA, models form also part of the discussion. This is a chapter, which is aimed at readers who are interested in understanding the technical background used in deriving results later in the book. 2.2. Summary Statistics for a set of Data A set of numbers which are generated by agents' actions in the market, and which can take different values if the agents were to repeat their actions, are known as random variables. For example, prices of shares in the stock exchange are determined at each point in time from the collective actions of agents operating in the market through their demand and supply decisions. The price of a share is thus a random variable, and so is the return of the share.

This book undertakes industry analysis of the risk-return behaviour of stock returns in the US. It focuses on transportation and a number of non-transportation companies, with particular attention paid to water transportation. The market risk (measured by beta) is identified as the relevant metric to use for comparison. Single and multi-index models are employed to identify exposures (betas) to market, microeconomic and macroeconomic types of risk in each industry. These industries include water transportation, air transportation, trucks, rail transportation, electricity, gas, petroleum refining and real estate. Microeconomic factors employed include market value of equity, book to market value of equity ratio, earnings to price ratio, asset to market value of equity ratio, and asset to book value of equity ratio. Macroeconomic factors include: industrial production, term structure of interest rates, oil prices, consumption and inflation. Returns of each industry seem to be explained by different sets of economic factors. For shipping, apart from the market, industrial production, oil prices, the market value of equity, and the asset to book ratio are significant.Moreover, the systematic risk of this industry is not different to that of the market, has not changed significantly under different market conditions, is lower than rail transportation, higher than real estate and similar to that of the remaining industries. The analysis is also extended beyond the US, to global industry portfolios.

The type of research methodologies used in analysing stock returns in the book is outlined in this technical chapter. The chapter begins with definitions of useful summary statistics, such as the mean, standard deviation, coefficient of variation, and their interpretation. Tests statistics for inferences on population means, proportions and differences of means, among others, are also presented. A summary of single equation regression techniques and the way these are applied through estimation and inference are outlined, focusing on the interpretation of standard output and diagnostic tests. The single equation methodology is then extended to multiple equation systems; the methods of Multivariate Least Square (MLSQ) of Seemingly Unrelated Regression Equations (SURE) are Regressions as well as those considered. Time series, ARlMA, models form also part of the discussion. This is a chapter, which is aimed at readers who are interested in understanding the technical background used in deriving results later in the book. 2.2. Summary Statistics for a set of Data A set of numbers which are generated by agents'' actions in the market, and which can take different values if the agents were to repeat their actions, are known as random variables. For example, prices of shares in the stock exchange are determined at each point in time from the collective actions of agents operating in the market through their demand and supply decisions. The price of a share is thus a random variable, and so is the return of the share.