Rangsorolás páros összehasonlításokkal. Kiegészítések a felvételizői preferencia-sorrendek módszertanához (Paired comparisons ranking. A supplement to the methodology of application-based preference ordering)

A Közgazdasági Szemle márciusi számában Telcs és szerzőtársai [2013] a felvételizők preferenciái alapján új megközelítést javasolt a felsőoktatási intézmények rangsorolására. Az alábbi írás új szempontokat biztosít ezen alapötlet gyakorlati megvalósításához. Megmutatja, hogy az alkalmazott modell ekvivalens az alternatívák egy aggregált páros összehasonlítási mátrix révén végzett rangsorolásával, ami rávilágít a szerzők kiinduló hipotéziseinek vitatható pontjaira. A szerző röviden áttekinti a hasonló feladatok megoldására javasolt módszereket, különös tekintettel azok axiomatikus megalapozására, majd megvizsgálja a Telcs és szerzőtársai [2013] által alkalmazott eljárásokat. Végül említést tesz egy hasonló megközelítéssel élő cikkről, és megfogalmaz néhány, a vizsgálat továbbfejlesztésére vonatkozó javaslatot. _____ In the March issue of Közgazdasági Szemle, Telcs et al. suggested a new approach to university ranking through preference ordering of applicants. The paper proposes new aspects to the implementation of this idea. It is shown that the model of these is equivalent to the ranking of alternatives based on paired comparisons, which reveals the debatable points in their hypotheses. The author reviews briefly the methods proposed in the literature, focusing on their axiomatic properties, and thoroughly examines the procedures of Telcs et al. [2013]. The paper presents an article which applied a similar approach and suggests some improvements to it.

Non-manipulable domains for the Borda count

We characterize the preference domains on which the Borda count satises Arrow's "independence of irrelevant alternatives" condition. Under a weak richness condition, these domains are obtained by xing one preference ordering and including all its cyclic permutations ("Condorcet cycles"). We then ask on which domains the Borda count is non-manipulable. It turns out that it is non-manipulable on a broader class of domains when combined with appropriately chosen tie-breaking rules. On the other hand, we also prove that the rich domains on which the Borda count is non-manipulable for all possible tie-breaking rules are again the cyclic permutation domains.

Nash implementable domains for the Borda count

We characterize the preference domains on which the Borda count satisfies Maskin monotonicity. The basic concept is the notion of a "cyclic permutation domain" which arises by fixing one particular ordering of alternatives and including all its cyclic permutations. The cyclic permutation domains are exactly the maximal domains on which the Borda count is strategy-proof when combined with every possible tie breaking rule. It turns out that the Borda count is monotonic on a larger class of domains. We show that the maximal domains on which the Borda count satisfies Maskin monotonicity are the "cyclically nested permutation domains" which are obtained from the cyclic permutation domains in an appropriately specified recursive way. ------ *We thank József Mala for posing the question of Nash implementability on restricted domains that led to this research. We are very grateful to two anonymous referees and an associate editor for their helpful comments and suggestions. The second author gratefully acknowledges financial support from the Hungarian Academy of Sciences (MTA) through the Bolyai János research fellowship.

Using the stated preference method for the calculation of social discount rate

The aim of this paper is to build the stated preference method into the social discount rate methodology. The first part of the paper presents the results of a survey about stated time preferences through pair-choice decision situations for various topics and time horizons. It is assumed that stated time preferences differ from calculated time preferences and that the extent of stated rates depends on the time period, and on how much respondents are financially and emotionally involved in the transactions. A significant question remains: how can the gap between the calculation and the results of surveys be resolved, and how can the real time preferences of individuals be interpreted using a social time preference rate. The second part of the paper estimates the social time preference rate for Hungary using the results of the survey, while paying special attention to the pure time preference component. The results suggest that the current method of calculation of the pure time preference rate does not reflect the real attitudes of individuals towards future generations.

Additive and multiplicative properties of scoring methods for preference aggregation

The paper reviews some additive and multiplicative properties of ranking procedures used for generalized tournaments with missing values and multiple comparisons. The methods analysed are the score, generalised row sum and least squares as well as fair bets and its variants. It is argued that generalised row sum should be applied not with a fixed parameter, but a variable one proportional to the number of known comparisons. It is shown that a natural additive property has strong links to independence of irrelevant matches, an axiom judged unfavourable when players have different opponents.

On the additivity of preference aggregation methods

The paper reviews some axioms of additivity concerning ranking methods used for generalized tournaments with possible missing values and multiple comparisons. It is shown that one of the most natural properties, called consistency, has strong links to independence of irrelevant comparisons, an axiom judged unfavourable when players have different opponents. Therefore some directions of weakening consistency are suggested, and several ranking methods, the score, generalized row sum and least squares as well as fair bets and its two variants (one of them entirely new) are analysed whether they satisfy the properties discussed. It turns out that least squares and generalized row sum with an appropriate parameter choice preserve the relative ranking of two objects if the ranking problems added have the same comparison structure.