Implicit rating: A potential new method to alert crisis on the interbank lending market

A new measure called “implicit rating” is introduced which might be a component of an early warning system. The proposed methodology relies on the aggregation of experts’ knowledge hidden in the transactional data of the interbank market of unsecured loans. Banks are simultaneously assessing the creditworthiness of each other which is reflected in the partner limits and in the interest rates. In the Hungarian interbank market the overall trading volume and the average interest rate did not show any negative trends before the crisis of 2008, however the average implicit partner limit started to decrease several months earlier, hence it might serve as a stress indicator.

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.