This is the story of the Princeton Wine Group, a group whose membership has been relatively constant for almost 40 years. This group has enjoyed 244 blind tastings involving 1,708 different wines. A statistical analysis was performed at each tasting examining whether participants ranked the quality of wines similarly and whether the preferences of the group were correlated with several variables including professional wine ratings and the prices of the wine. The article concludes with a discussion of lessons learned from a lifetime of wine tastings.

Compositional data for 464 clay minerals (2:1 type) were analyzed by statistical techniques. The objective was to understand the similarities and differences between the groups and subgroups and to evaluate statistically clay mineral classification in terms of chemical parameters. The statistical properties of the distributions of total layer charge (TLC), K, VIAl, VIMg, octahedral charge (OC) and tetrahedral charge (TC) were initially evaluated. Critical-difference (P = 1%) comparisons of individual characteristics show that all the clay micas (illite, glauconite and celadonite) differ significantly from all the smectites (montmorillonite, beidellite, nontronite and saponite) only in their TLC and K levels; they cannot be distinguished by their VIAl, VIMg, TC or OC values which reveal no significant differences between several minerals.

Linear discriminant analysis using equal prior was therefore performed to analyze the combined effect of all the chemical parameters. Using six parameters [TLC, K, VIAl, VIMg, TC and OC], eight minerals groups could be derived, corresponding to the three clay micas, four smectites (mentioned above) and vermiculite. The fit between predicted and experimental values was 88.1%. Discriminant analysis using two parameters (TLC and K) resulted in classification into three broad groups corresponding to the clay micas, smectites and vermiculites (87.7% fit). Further analysis using the remaining four parameters resulted in subgroup-level classification with an 85–95% fit between predicted and experimental results. The three analyses yielded D2 Mahalanobis distances, which quantify chemical similarities and differences between the broad groups, within members of a subgroup and also between the subgroups. Classification functions derived here can be used as an aid for classification of 2:1 minerals.

Subcutaneous fat thickness and fatty acid composition (FAC) play an important role on seasoning loss and organoleptic characteristics of seasoned hams. Dry-cured ham industry prefers meats with low contents of polyunsaturated fatty acids (PUFA) because these negatively affect fat firmness and ham quality, whereas consumers require higher contents in those fatty acids (FA) for their positive effect on human health. A population of 950 Italian Large White pigs from the Italian National Sib Test Selection Programme was investigated with the aim to estimate heritabilities, genetic and phenotypic correlations of backfat FAC, Semimembranosus muscle intramuscular fat (IMF) content and other carcass traits. The pigs were reared in controlled environmental condition at the same central testing station and were slaughtered at reaching 150 kg live weight. Backfat samples were collected to analyze FAC by gas chromatography. Carcass traits showed heritability levels from 0.087 for estimated carcass lean percentage to 0.361 for hot carcass weight. Heritability values of FA classes were low-to-moderate, all in the range 0.245 for n-3 PUFA to 0.264 for monounsaturated FA (MUFA). Polyunsaturated fatty acids showed a significant genetic correlation with loin thickness (0.128), backfat thickness (−0.124 for backfat measured by Fat-O-Meat’er and −0.175 for backfat measured by calibre) and IMF (−0.102). Obviously, C18:2(n-6) shows similar genetic correlations with the same traits (0.211 with loin thickness, −0.206 with backfat measured by Fat-O-Meat’er, −0.291 with backfat measured by calibre and −0.171 with IMF). Monounsaturated FA, except with the backfat measured by calibre (0.068; P<0.01), do not show genetic correlations with carcass characteristics, whereas a negative genetic correlation was found between MUFA and saturated FA (SFA; −0.339; P<0.001). These results suggest that MUFA/SFA ratio could be increased without interfering with carcass traits. The level of genetic correlations between FA and carcass traits should be taken into account in dealing with the development of selection schemes addressed to modify carcass composition and/or backfat FAC.

We give an asymptotic formula for correlations

$$\begin{eqnarray}\mathop{\sum }_{n\leqslant x}f_{1}(P_{1}(n))f_{2}(P_{2}(n))\cdots f_{m}(P_{m}(n)),\end{eqnarray}$$

$f,\ldots ,f_{m}$

$f:\mathbb{N}\rightarrow \{-1,+1\}$

$1957$

$f(n)=n^{s}$

$\operatorname{Re}(s)<1$

$|f(n)|$

$n=a+b$

$a,b$

$\mathbb{N}$

This paper seeks to establish good practice in setting inputs for operational risk models for banks, insurers and other financial service firms. It reviews Basel, Solvency II and other regulatory requirements as well as publicly available literature on operational risk modelling. It recommends a combination of historic loss data and scenario analysis for modelling of individual risks, setting out issues with these data, and outlining good practice for loss data collection and scenario analysis. It recommends the use of expert judgement for setting correlations, and addresses information requirements for risk mitigation allowances and capital allocation, before briefly covering Bayesian network methods for modelling operational risks.

The difficulties and costs of measuring individual feed intake in dairy cattle are the primary factors limiting the genetic study of feed intake and utilisation, and hence the potential of their subsequent industry-wide applications. However, indirect selection based on heritable, easily measurable, and genetically correlated traits, such as conformation traits, may be an alternative approach to improve feed efficiency. The aim of this study was to estimate genetic and phenotypic correlations among feed intake, production, and feed efficiency traits (particularly residual feed intake; RFI) with routinely recorded conformation traits. A total of 496 repeated records from 260 Holstein dairy cows in different lactations (260, 159 and 77 from first, second and third lactation, respectively) were considered in this study. Individual daily feed intake and monthly BW and body condition scores of these animals were recorded from 5 to 305 days in milk within each lactation from June 2007 to July 2013. Milk yield and composition data of all animals within each lactation were retrieved, and the first lactation conformation traits for primiparous animals were extracted from databases. Individual RFI over 301 days was estimated using linear regression of total 301 days actual energy intake on a total of 301 days estimated traits of metabolic BW, milk production energy requirement, and empty BW change. Pair-wise bivariate animal models were used to estimate genetic and phenotypic parameters among the studied traits. Estimated heritabilities of total intake and production traits ranged from 0.27±0.07 for lactation actual energy intake to 0.45±0.08 for average body condition score over 301 days of the lactation period. RFI showed a moderate heritability estimate (0.20±0.03) and non-significant phenotypic and genetic correlations with lactation 3.5 % fat-corrected milk and average BW over lactation. Among the conformation traits, dairy strength, stature, rear attachment width, chest width and pin width had significant (P<0.05) moderate to strong genetic correlations with RFI. Combinations of these conformation traits could be used as RFI indicators in the dairy genetic improvement programmes to increase the accuracy of the genetic evaluation of feed intake and utilisation included in the index.

Examples of distinct weighted model sets with equal 2, 3, 4, 5-point correlations are given.

The long-chain n-3 PUFA, EPA, is believed to be important for skin health, including roles in the modulation of inflammation and protection from photodamage. FFQ and blood levels are used as non-invasive proxies for assessing skin PUFA levels, but studies examining how well these proxies reflect target organ content are lacking. In seventy-eight healthy women (mean age 42·8, range 21–60 years) residing in Greater Manchester, we performed a quantitative analysis of long-chain n-3 PUFA nutrition estimated from a self-reported FFQ (n 75) and correlated this with n-3 PUFA concentrations in erythrocytes (n 72) and dermis (n 39). Linear associations between the three n-3 PUFA measurements were assessed by Spearman correlation coefficients and agreement between these measurements was estimated. Average total dietary content of the principal long-chain n-3 PUFA EPA and DHA was 171 (sd 168) and 236 (sd 248) mg/d, respectively. EPA showed significant correlations between FFQ assessments and both erythrocyte (r 0·57, P< 0·0001) and dermal (r 0·33, P= 0·05) levels, as well as between erythrocytes and dermis (r 0·45, P= 0·008). FFQ intake of DHA and the sum of n-3 PUFA also correlated well with erythrocyte concentrations (r 0·50, P< 0·0001; r 0·27, P= 0·03). Agreement between ranked thirds of dietary intake, blood and dermis approached 50 % for EPA and DHA, though gross misclassification was lower for EPA. Thus, FFQ estimates and circulating levels of the dietary long-chain n-3 PUFA, EPA, may be utilised as well-correlated measures of its dermal bioavailability.

This paper proposes a multivariate unobserved-components model to simultaneously decompose the real GDP for each of the G-7 countries into its respective trend and cycle components. In contrast to previous literature, our model allows for explicit correlation between all the contemporaneous trend and cycle shocks. We find that all the G-7 countries have highly variable stochastic permanent components for output, even once we allow for structural breaks. We also find that common restrictions on the correlations between trend and cycle shocks are rejected by the data. In particular, we find that correlations across permanent and transitory shocks are important both within and across countries.

The objectives of the present study were to identify dietary patterns independently in first-time mothers and fathers, and to examine whether these patterns were correlated within families. Dietary intakes were collected at baseline in the Melbourne Infant Feeding Activity and Nutrition Trial Program using a validated FFQ in 454 pairs of first-time mothers and fathers. Education level was reported in associated questionnaires. Principal components analyses included frequencies of fifty-five food groups and were performed independently in mothers and fathers. Spearman's correlation coefficients were used to assess associations between dietary pattern scores. A total of four dietary patterns were identified in mothers and fathers. Of these, three dietary patterns had similar characteristics between these two populations, namely ‘Fruits and vegetables’, ‘High-energy snack and processed foods’, ‘High-fat foods’ in mothers; and ‘Fruits’, ‘High-energy snack and processed foods’, ‘High-fat foods’ in fathers. The following two additional patterns were identified: ‘Cereals and sweet foods’ in mothers and ‘Potatoes and vegetables’ in fathers. Patterns incorporating healthier food items were found to be positively associated with parent education. An inverse association with education was found for the ‘High-fat foods’ and ‘High-energy snack and processed foods’ dietary patterns. Qualitatively similar patterns between corresponding mothers and fathers were the most strongly correlated (ρ = 0·34–0·45, P < 0·001). There were some differences in dietary patterns between mothers and fathers, suggesting that it is worth deriving patterns separately when considering couples, and more generally between men and women. Exploring how these various patterns correlate within households provides important insights to guide the development and implementation of family-based interventions.