De jury heeft de volgende inzendingen ontvangen.
Deelnemer: Julia Mensink
Titel scriptie: Index numbers: from artefacts to standards
Index numbers make it possible to compare between different regions and different periods. It is a way to look at economic reality by manipulating it into numbers. How do index numbers represent the world? Are they accurate and precise? This deals with their construction, the steps undertaken from the world to the resulting index. The term artifact is relevant here. The term ‘artifact’ can have two meanings, the first being ‘distinct from natural’, and the second being ‘distinct from fact’. That is, an artifact is caused by human intervention, thus artificial as opposed to natural. But as the term is used in medicine, it can also be an outcome not inherent in the natural state of the object studied, but occurring because of the manner of investigation. In this meaning an artifact is not a fact but a spurious outcome.
After construction an index can start to lead a life of its own. This development is a different process than the construction. Starting off being an artifact, it can end up being widely used in academic and policy settings. What makes this development occur? And further, how are the index numbers used? The settings are interrelated but also clearly have differences.
Since 1990 the United Nations Development Programme (UNDP) has published the Human Development Report (HDR) in which a wide range of indicators give an idea of the world’s development. The main eye catcher is the Human Development Index (HDI). It shows human development in a digit between zero and one where a higher score means higher development. It is constructed by taking the simple average of three aspects: longevity, knowledge, and a decent standard of living.
In this thesis the construction, development and usage of index numbers will be examined, focusing on the Human Development Index. Issues as quantification, construction of index numbers, development of an index- from artifact to standard- and the usage in academic and policy setting will be discussed.
Deelnemer: Michiel Nijhuis
Titel scriptie: Method validation: confidence intervals on measures of precision
Within the pharmaceutical industry it is common practice to transfer analytical methods between laboratories. From these interlaboratory studies the repeatability, the intermediate precision and the reproducibility of the analytical method are calculated. These measures are appropriate sums of variance components from an analysis of variance model describing the structure of the data. In the literature several methods have been described for calculating (approximate) confidence intervals on sums of variance components, i.e. Satterthwaite, Welch and Modified Large-Sample. Comparison studies between these methods have been performed for one or two-way classification analysis of variance models only. Interlaboratory studies often need higher order classifications. We will discuss one specific three-way classification analysis of variance model for the transfer of analytical methods that is used frequently. In practice one or more variance components may be estimated negatively. These negative estimates can be handled in several ways. For instance these values can be used in the calculation of the measures of precision or these values can be set equal to zero before the measures of precision are calculated. By means of a simulation study the performance of the different approximation methods was compared for both calculation procedures.
Deelnemer: Mariëlle Non
Titel scriptie: Response Styles: Occurrence and Correction Methods
The topic of this thesis is response styles. Response styles are defined as a tendency to distort responses in a particular direction regardless of the content of the question. The main focus of response styles research is on Likert-type of questions, where people are asked to give an answer on a numerical scale. For example, people who suffer from a yea-saying/nay-saying response style always give too high or too low answers and people who have a response range bias have a tendency to answer in a too narrow or too wide range around the mean response. In this thesis I use a questionaire where people are asked for the probability of several events to happen. I first show that the answers suffer from yea-saying and response range bias. Then I try several correction methods to correct for these response styles and some other response styles as well. Most of the correction methods lead to a significantly better predictive value of the answers of the respondents, while the correction method correcting for all types of response styles considered in this thesis generally performs best. A point of attention is the way how response styles are measured: different measurement methods lead to significantly different results, although the theory expects no influence of measurement method.
Deelnemer: Marnix Engels
Titel scriptie: Portfolio Optimization: Beyond Markowitz
Every investor is looking for an optimal portfolio of assets that gives a high return with low risk. But what is risk? In 1952, Harry Markowitz wrote a paper about modern portfolio theory where he defined the standard deviation of a portfolio as the risk measure. He supposed that if an investor can choose between two portfolios with the same expected return, he will choose the one with the lowest standard deviation. Markowitz got the Nobel price in 1990 for his theory.
This thesis tries to go beyond the theory of Markowitz. There are more definitions of risk, like the "safety first" principle. Not the standard deviation, but shortfall probabilities are the risk measure. Basic thought is that an investor is interested in finding a maximum expected return with a low chance of losing a certain amount of money. A wide overview of the differences between and similarities of the two models is given.
The thesis further discusses what happens if returns are supposed to be elliptically distributed instead of normally. Also an overview of the widely used risk measure "Value at Risk" is given and other performance measures than the expected return are used. Furthermore two proposals of dealing with parameter uncertainty are done.