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.

On the sub-Gaussianity of the Beta and Dirichlet distributions

Abstract: We obtain the optimal proxy variance for the sub-Gaussianity of Beta distribution, thus proving upper bounds recently conjectured by Elder (2016). We provide different proof techniques for the symmetrical (around its mean) case and the non-symmetrical case. The technique in the latter case relies on studying the ordinary differential equation satisfied by the Beta moment-generating function known as the confluent hypergeometric function. As a consequence, we derive the optimal proxy variance for the Dirichlet distribution, which is apparently a novel result. We also provide a new proof of the optimal proxy variance for the Bernoulli distribution, and discuss in this context the proxy variance relation to log-Sobolev inequalities and transport inequalities.

Preferring proxy-agency: impact on self-efficacy for exercise.

Abstract: We examined: (a) the social-cognitions of those who differ in preferred level of proxy-assistance in both proxy-led and independent exercise contexts; and (b) the relationships between proxy-agency and reliance Sixty-five fitness class participants completed measures of self-regulatory efficacy, task self-efficacy, exercise intentions, proxy-efficacy and perceived reliance A 2 (exercise context) by 2 (preferred assistance) MANOVA revealed significant main effects for assistance and context Also, significant assistance by context interactions was found for both efficacies When facing exercise without a proxy, individuals preferring high proxy assistance expressed lower self-regulatory and task self-efficacy Finally, proxy- and self-regulatory efficacy were shown to be separate predictors of reliance on a proxy Results support Bandura's theorizing about the proxy-led context and its influence on self-regulatory efficacy

The Disciplinary Effects of Proxy Contests

Abstract: Using a manually collected data set of all proxy contests from 1994 through 2012, I show that proxy contests play an important role in hostile corporate governance. Target shareholders benefit from proxy contests: the average abnormal returns reach 6.5% around proxy contest announcements. Proxy contests that address firms' business strategies and undervaluation are most beneficial for shareholders. In contrast, proxy contests that aim at changing capital structure and governance do not lead to higher firm values. Relative to matching firms, future targets are smaller, they have higher stock liquidity, higher institutional and activist ownership, lower leverage and market valuation, and higher investments. While most of these characteristics predict proxy contests in time-series, prior to proxy contests targets also experience poor stock performance, decreases in investments, increases in cash reserves and payouts to shareholders, and increases in management's entrenchment. These changes in corporate policies are consistent with targets' attempts to affect the probability of a proxy contest.

Inter-rater agreement of the Quality of Life-Alzheimer's Disease (QoL-AD) self-rating and proxy rating scale : secondary analysis of RightTimePlaceCare data

Abstract: To assess the quality of life of people with dementia, measures are required for self-rating by the person with dementia, and for proxy rating by others. The Quality of Life in Alzheimer’s Disease scale (QoL-AD) is available in two versions, QoL-AD-SR (self-rating) and QoL-AD-PR (proxy rating). The aim of our study was to analyse the inter-rater agreement between self- and proxy ratings, in terms of both the total score and the items, including an analysis specific to care setting, and to identify factors associated with this agreement. Cross-sectional QoL-AD data from the 7th Framework European RightTimePlaceCare study were analysed. A total of 1330 cases were included: n = 854 receiving home care and n = 476 receiving institutional long-term nursing care. The proxy raters were informal carers (home care) and best-informed professional carers (institutional long-term nursing care). Inter-rater agreement was investigated using Bland-Altman plots for the QoL-AD total score and by weighted kappa statistics for single items. Associations were investigated by regression analysis. The overall QoL-AD assessment of those with dementia revealed a mean value of 33.2 points, and the proxy ratings revealed a mean value of 29.8 points. The Bland-Altman plots revealed a poor agreement between self- and proxy ratings for the overall sample and for both care settings. With one exception (item ‘Marriage’ weighted kappa 0.26), the weighted kappa values for the single QoL-AD items were below 0.20, indicating poor agreement. Home care setting, dementia-related behavioural and psychological symptoms, and the functional status of the person with dementia, along with the caregiver burden, were associated with the level of agreement. Only the home care setting was associated with an increase larger than the predefined acceptable difference between self- and proxy ratings. Proxy quality of life ratings from professional and informal carers appear to be lower than the self-ratings of those with dementia. QoL-AD-SR and QoL-AD-PR are therefore not interchangeable, as the inter-rater agreement differs distinctly. Thus, a proxy rating should be judged as a complementary perspective for a self-assessment of quality of life by those with dementia, rather than as a valid substitute.