There is a great need for robust techniques in data mining and machine learning contexts where many standard techniques such as principal component analysis and linear discriminant analysis are inherently susceptible to outliers. Furthermore, standard robust procedures assume that less than half the observation rows of a data matrix are contaminated, which may not be a realistic assumption when the number of observed features is large. This work looks at the problem of estimating covariance and precision matrices under cellwise contamination. We consider using a robust pairwise covariance matrix as an input to various regularisation routines, such as the graphical lasso, QUIC and CLIME. To ensure the input covariance matrix is positive semidefinite, we use a method that transforms a symmetric matrix of pairwise covariances to the nearest covariance matrix. The result is a potentially sparse precision matrix that is resilient to moderate levels of cellwise contamination. Since this procedure is not based on subsampling it scales well as the number of variables increases.

No abstract

Simple SummaryProducing a product that delivers a consistently high-quality eating experience is paramount to the Australian beef industry to ensure consumer satisfaction and return protein purchasing. The importance of minimising pre-slaughter stress in cattle for animal welfare and meat quality is well understood by the industry, however, there currently exists no objective measurement of detecting which cattle are at greatest risk of producing poor quality meat. A pre-slaughter measurement would enable the beef industry to detect at risk cattle and implement an intervention strategy prior to slaughter. Muscle damage enzyme creatine kinase was the plasma biomarker most correlated with meat quality and a two-week rest period prior to slaughter was beneficial for improving quality. Further research is required to determine the usefulness of creatine kinase as an objective measurement on a commercial scale and the cost benefit of a two-week rest period for the industry. AbstractThis study considered the relationship between pre-slaughter stressors and plasma biomarkers in 488 pasture-raised cattle across two experiments. The design aimed to test groups consisting of steer only, heifer only, and mixed sex cattle under direct kill versus rested (14 days in abattoir holding paddocks) protocols. In Experiment One, cattle were sourced from four farms, and transported by trucks and ships on the same day. In Experiment Two, cattle were sourced from four farms where a comparison was made between marketing via two commercial saleyards or direct farm gate consignment to abattoir. Blood samples were collected at exsanguination for subsequent analyses and relation to meat quality attributes. Muscle damage, as indicated by creatine kinase, is the biomarker most correlated to ultimate pH and muscle glycogen concentrations. A two-week rest period is effective for lowering this enzyme and improving muscle glycogen concentration. Although the cattle was subjected to a range of stress inducing treatments, we found that plasma biomarkers alone appeared insufficient for use as diagnostic stress indicators.

As marbling is a principal input into many grading systems it is important to have an accurate and reliable measurement procedure. This paper compares three approaches to measuring marbling: trained personnel, near infrared spectroscopy (NIR) and image analysis. One 25mm slice of meat was utilised from up to 12 cuts from 48 carcasses processed in Poland and France. Each slice was frozen to enable a consistent post-slaughter period then thawed for image analysis. The images were appraised by experienced beef graders and the sample used to determine fat content by NIR. We find that image analysis based marbling measures are capturing something different to trained personnel and that there is a strong relationship between near infrared spectroscopy and trained personnel. Finally, we demonstrate that marbling measures taken on one muscle can be predictive of marbling in other muscles in the same carcase. This is particularly important for cut based models such as the Meat Standards Australia system.

No abstract

Simple SummaryConsumer satisfaction is essential to the beef industry to ensure return protein purchasing. The Meat Standards Australia grading system has been pivotal in regulating the quality and consistency of meat palatability by creating objective measurements such as carcass characteristics, genetics systems, and production systems to predict consumer eating quality. One of the main objective measurements for carcass compliance is the ultimate pH of the longissimus thoracis et lumborum; however, recent research has demonstrated that pre-slaughter stress is eroding beef eating quality in pH compliant carcasses. Pre-slaughter mixing and transport was associated with lower eating quality in certain cuts. A two-week rest period at the abattoir prior to slaughter was beneficial for improving consumer sensory eating quality scores. Further research is required to determine if the muscle damage enzyme creatine kinase could be used commercially as an objective measurement to identify which cattle could benefit from a pre-slaughter rest period to improve beef quality.AbstractThe study evaluated the relationship between pre-slaughter stress, plasma biomarkers and consumer-evaluated eating quality of pasture raised beef cattle (n = 488). The design tested steer only, heifer only and mixed sex cattle with a comparison of direct kill versus a 14 day rest period in abattoir holding paddocks prior to slaughter. Experiment One sourced cattle from four farms and tested shipping and road transport. Experiment Two sourced cattle from four farms and tested a commercial saleyard pathway. The impact on treatment on untrained consumer eating quality scores were tested on five muscle groups, m. psoas major, m. longissimus dorsi lumborum, m. biceps femoris, m. semitendinosis, and m. infraspinatus. Across all muscles, a two-week rest period had the biggest improvement in sensory score. Mixed groups scored lower in the outside muscle than non-mixed groups. However, the mixing response was inconsistent in the eye round muscle and not significant in the other muscles. Plasma glucose and L-lactate indicated a marked acute stress response at slaughter with a small detrimental impact on consumer score. The muscle damage enzyme markers creatine kinase (CK) and aspartate aminotransferase (AST) were strongly associated with a lower meat quality score (MQ4). Neither β-hydroxybutyrate (βHB) nor non-esterified fatty acids (NEFA) were associated with MQ4, suggesting that fat mobilisation does not impact consumer sensory score.

Since the introduction of regression quantiles for estimating conditional quantile functions there has been ongoing research into how best to construct confidence intervals for parameter estimates. The three main methods are direct estimation, rank test inversion and resampling methods. Kocherginsky et al. [Practical confidence intervals for regression quantiles, J. Comput. Graph. Statist. 14 (2005), pp. 41-55] gave an overview of some of the available procedures. Five years on, the aim of this paper is to revisit and extend their analysis, evaluating additional techniques with a focus on smaller sample sizes and more extreme conditional quantiles. In particular, we find the percentile bootstrap (pbs) to be an eminently viable alternative for confidence interval construction. We show that it provides empirical coverage probabilities generally as good as, or better than, the other more complex resampling methods. Furthermore, pbs confidence intervals typically exhibit smaller average lengths across a variety of models than those based on the rank inversion methods which, like the pbs, avoids explicitly estimating asymptotic variances.

No abstract