Topic
Proxy (statistics)
About: Proxy (statistics) is a research topic. Over the lifetime, 5257 publications have been published within this topic receiving 94504 citations. The topic is also known as: proxy variable & proxy measurement.
Papers published on a yearly basis
Papers
More filters
Patent•
25 Nov 1997TL;DR: In this article, a set of proxy servers are dynamically identified by a manager processor in accordance with an invention, and the manager processor also helps search the prefetched objects for the requested object.
Abstract: In a communication arrangement including a multiplicity of intranets, objects available on the Internet are requested. A requested object in an intranet is attended by a proxy server therein, which attempts to locate a copy of the requested object in a proxy cache. The latter stores copies of objects previously obtained by web browsers in the intranet, and affords priority to copies of those objects from selected Internet sites. If the proxy server does not manage to locate a copy of the requested object or determines that a located copy is outdated, the proxy server seeks cooperation from other proxy servers to further search for the object. These cooperating proxy servers are dynamically identified by a manager processor in accordance with the invention. In addition, the manager processor prefetches selected objects from the Internet in anticipation of the object request. The manager processor also helps search the prefetched objects for the requested object.
33 citations
TL;DR: In this paper, the problem of calibrating the CreditMetricsTM correlation concept for a portfolio of loans to medium-sized, non-listed German firms has been addressed and the solution to this estimation problem offered in CreditManagerTM, which basically relates the weight of the idiosyncratic component to company size, can be adapted to meet the needs of their application.
Abstract: This paper deals with the problems associated with empirically calibrating the CreditMetricsTM correlation concept for a portfolio of loans to medium-sized, non-listed German firms. Using this framework to determine the probabilities of joint default events, requires an estimation of the portion of the obligors' asset return volatility that is firm-specific (idiosyncratic). We analyze if and how the solution to this estimation problem offered in CreditManagerTM, which basically relates the weight of the idiosyncratic component to company size, can be adapted to meet the needs of our application. Using a random sample of 250 German stocks, we show that there is indeed a significant positive relationship between R-squared and market capitalization (book value of total assets). However, this relationship appears to hold only for stocks with considerable index weights. For "smaller" firms, both size proxies perform very poorly. We also test the explanatory power of four different versions of the CreditMetricsTM index model to predict asset correlations. While the use of the book value of total assets as a proxy for size does not contribute to explaining obligors' pairwise asset correlations beyond the correlations of their corresponding CDAX(R) industry indices at all, the use of market capitalization does - but only with respect to "bigger" companies.
33 citations
TL;DR: A large body of work in the social sciences relies on proxy variables to capture the influence of an unobserved regressor as mentioned in this paper, assuming that measurement error is well approximated by a classical model.
Abstract: A large body of work in the social sciences relies on proxy variables to capture the influence of an unobserved regressor. Assuming that measurement error is well approximated by a classical model ...
33 citations
01 Mar 1996
TL;DR: This document explains "classical" and "transparent" proxy techniques and attempts to provide rules to help determine when each proxy system may be used without causing problems.
Abstract: Many modern IP security systems (also called "firewalls" in the trade) make use of proxy technology to achieve access control. This document explains "classical" and "transparent" proxy techniques and attempts to provide rules to help determine when each proxy system may be used without causing problems.
33 citations
TL;DR: This paper found that institutional investors and proxy advisors pay attention to the changing opinions of their beneficiaries and shareholders, as reflected in their voting decisions, and that the proxy voting process serves as a channel for the public to influence corporate behavior.
Abstract: Institutional investors vote corporate proxies on behalf of underlying investors and beneficiaries. We show a strong relation between this voting and public opinion on corporate governance (as reflected in media coverage and surveys), with similarly strong results for voting by mutual funds. We also find that proxy advisors’ recommendations are associated with public opinion. Our results suggest that institutional investors and proxy advisors pay attention to the changing opinions of their beneficiaries and shareholders, as reflected in their voting decisions, and that the proxy voting process serves as a channel for the public to influence corporate behavior.
33 citations