Incorporating Higher Moments into Value at Risk Forecasting

Value-at-risk (VaR) forecasting generally relies on a parametric density function of portfolio returns that ignores higher moments or assumes them constant. In this paper, we propose a simple approach to forecasting of a portfolio VaR. We employ the Gram-Charlier expansion (GCE) augmenting the standard normal distribution with the first four moments, which are allowed to vary over time. In an extensive empirical study, we compare the GCE approach to other models of VaR forecasting and conclude that it provides accurate and robust estimates of the realized VaR. In spite of its simplicity, on our dataset GCE outperforms other estimates that are generated by both constant and time-varying higher-moments models.

ESC observations of SN 2005cf - I. Photometric evolution of a normal Type Ia supernova

We present early-time optical and near-infrared photometry of supernova (SN) 2005cf. The observations, spanning a period from about 12 d before to 3 months after maximum, have been obtained through the coordination of observational efforts of various nodes of the European Supernova Collaboration and including data obtained at the 2-m Himalayan Chandra Telescope. From the observed light curve we deduce that SN 2005cf is a fairly typical SN Ia with a post-maximum decline [Delta m(15)(B)(true) = 1.12] close to the average value and a normal luminosity of M-B,M-max = -19.39 +/- 0.33. Models of the bolometric light curve suggest a synthesized Ni-56 mass of about 0.7 M-circle dot. The negligible host galaxy interstellar extinction and its proximity make SN 2005cf a good Type Ia SN template.

The short-term effect of topical dorzolamide compared with placebo on perimetry testing in normal subjects

Purpose. To study the effect of topical dorzolamide on retinal sensitivity in normal subjects. Methods. 20 normal subjects were prospectively randomized in a two period crossover study. Baseline threshold perimetry (HVF, program 24-2, Statpac analysis) was obtained. The study eye was randomly chosen and then was randomly assigned to receive either dorzolamide 2% or placebo one hour before repeat HVF. After at least two weeks washout the same eye was given the opposite drop one hour prior to repeat HVF. Global indices were compared using a paired T-test. Results. AFTER PLACEBO AFTER DORZOLAMIDE P-VALUE (mean±SD) (mean±SD) MD 0.12±1.4 0.16±1.4 0.82 PSD 1.83±0.6 1.66±0.4 0.04 SF 1.28±0.4 1.24±0.2 0.73 CPSD 1.05±0.9 0.84±0.6 0.35 Conclusion. In this population, topical dorzolamide had little effect on visual function.

The prognostic and therapeutic value of EpHA2 in early colorectal cancer (CRC).

Background: EpHA2 is a 130 kD transmembrane glycoprotein belonging to ephrin receptor subfamily and involved in angiogenesis/tumour neovascularisation. High EpHA2 mRNA level has recently been implicated in cetuximab resistance. Previously, we found high EpHA2 levels in a panel of invasive colorectal cancer (CRC) cells, which was associated with high levels of stem-cell marker CD44. Our aim was to investigate the prognostic value of EpHA2 and subsequently correlate expression levels to known clinico-pathological variables in early stage CRC. Methods: Tissue samples from 509 CRC patients were analysed. EpHA2 expression was measured using IHC. Kaplan-Meier graphs were used. Univariate and multivariate analyses employed Cox Proportional Hazards Ratio (HR) method. A backward selection method (Akaike’s information criterion) was used to determine a refined multivariate model. Results: EpHA2 was highly expressed in CRC adenocarcinoma compared to matched normal colon tissue. In support of our preclinical invasive models, strong correlation was found between EpHA2 expression and CD44 and Lgr5 staining (p<0.001). In addition, high EpHA2 expression significantly correlated with vascular invasion (p=0.03).HR for OS for stage II/III patients with high EpHA2 expression was 1.69 (95%CI: 1.164-2.439; p=0.003). When stage II/III was broken down into individual stages, there was significant correlation between high EpHA2 expression and poor 5-years OS in stage II patients (HR: 2.18; 95%CI: 1.28-3.71; p=0.005).HR in the stage III group showed a trend to statistical significance (HR: 1.48; 95%CI=0.87-2.51; p=0.05). In both univariate and multivariate analyses of stage II patients, high EpHA2 expression was the only significant factor and was retained in the final multivariate model. Higher levels of EpHA2 were noted in our RAS and BRAF mutant CRC cells, and silencing EpHA2 resulted in significant decreases in migration/invasion in parental and invasive CRC sublines. Correlation between KRAS/NRAS/BRAFmutational status and EpHA2 expression in clinical samples is ongoing. Conclusions: Taken together, our study is the first to indicate that EpHA2 expression is a predictor of poor clinical outcome and a potential novel target in early stage CRC.