No abstract

The selection of feature genes with high recognition ability from the gene expression profiles has gained great significance in biology. However, most of the existing methods have a high time complexity and poor classification performance. Motivated by this, an effective feature selection method, called supervised locally linear embedding and Spearman's rank correlation coefficient (SLLE-SC2), is proposed which is based on the concept of locally linear embedding and correlation coefficient algorithms. Supervised locally linear embedding takes into account class label information and improves the classification performance. Furthermore, Spearman's rank correlation coefficient is used to remove the coexpression genes. The experiment results obtained on four public tumor microarray datasets illustrate that our method is valid and feasible.

Purpose This paper aims to investigate patterns in UK stock returns related to downside risk, with particular focus on stock returns during financial crises. Design/methodology/approach First, stocks are sorted into five quintile portfolios based on the relevant beta values (classic beta, downside beta and upside beta, calculated by the moving window approach). Second, patterns of portfolio returns are examined during various sub-periods. Finally, predictive powers of beta and downside beta are examined. Findings The downside risk is observed to have a significant positive impact on contemporaneous stock returns and a negative impact on future returns in general. In contrast, an inverse relationship between risk and return is observed when stocks are sorted by beta, contrary to the classic literature. UK stock returns exhibit clear time sensitivity, especially during financial crises. Originality/value This paper focuses on the impact of the downside risk on UK stock returns, assessed via a comprehensive sub-period analysis. This paper fills the gap in the existing literature, in which very few studies examine the time sensitivity in relation to the downside risk and the risk-return anomaly in the UK stock market using a long sample period.

No abstract

Background Caring duties fall disproportionally on particular demographic groups (eg, women and individuals aged 45 years or older) and can have adverse effects on their mental wellbeing. Effects on carers tend to worsen as the extent of care increases but how sociodemographic factors can strengthen or weaken the association is unclear. We aimed to assess the association between socioeconomic position and caring intensity on the mental wellbeing of unpaid carers. MethodsIn this cross-sectional study, we analysed data from unpaid carers (aged ≥16 years) using the 2016-17 National Survey for Wales (NSW). Mental wellbeing was measured using the Warwick-Edinburgh Mental Wellbeing Score (WEMWBS). NSW collected information on carers' demographics, caring intensity (caring hours per week) alongside with socioeconomic position (economic status, education, and income deprivation). Ordinary least square regression was used to examine how demographic factors modify the association between caring intensity and average mental wellbeing. We used quantile regression to examine how socioeconomic position contributes to specific percentile points of WEMWBS, with a focus on the pattern from low (15th percentile) to high (85th percentile) mental wellbeing. FindingsWe identified 2144 unpaid carers (40•3% men and 59•7% women) for analysis. Increasing caring intensity negatively correlated with mental wellbeing (high vs low intensity, estimated WEMWBS 49 vs 51, p=0•011) and the correlation exacerbated in carers who were female (48 vs 50, p=0•028) , age 65 years or older (49 vs 54, p=0•0001), and providing care while working (46 vs 49, p=0•014). Carers who were economically inactive (50 vs 52, p=0•0012) and with low level of education (ie, NFQ level 3 or below) (50 vs 52, p=0•0004) were more likely to experience poor mental wellbeing. A decreasing trend was observed in the differences between economically active and inactive carers' mental wellbeing from poor to good mental wellbeing percentiles (p=0•011). A similar pattern was observed in education (p=0•042), but not income deprivation. Interpretation Caring intensity is associated with mental wellbeing, but the effects vary by demographics. Future public health policy and intervention development should consider carers who are disproportionally affected by caring responsibilities and clarify whether the aim is to prevent poor mental wellbeing or promote mental wellbeing.Funding Public Health Wales.