Trends in the skewness of the body mass index distribution among urban Australian adults, 1980 to 2007

Purpose We analyzed the changes in the body mass index (BMI) distribution for urban Australian adults between 1980 and 2007.

Methods We used data from participants of six consecutive Australian nation-wide surveys with measured weight and height between 1980 and 2007. We used quantile regression to estimate mean BMI (for percentiles of BMI) and prevalence of severe obesity, modeled by natural splines in age, date of birth, and survey date.

Results Since 1980, the right skew in the BMI distribution for Australian adults has increased greatly for men and women, driven by increases in skew associated with age and birth cohort/period. Between 1980 and 2007, the average 5-year increase in BMI was 1 kg/m2 (0.8) for the 95th percentile of BMI in women (men). The increase in the median was about a third of this, and for the 10th percentile, a fifth of this. We estimated that for the cohort born in 1960 around 31% of men and women were obese by age 50 years compared with 11% of the 1930 birth cohort.

Conclusions There have been large increases in the right skew of the BMI distribution for urban Australian adults between 1980 and 2007, and birth cohort effects suggests similar increases are likely to continue.

Robustness to non-normality of various tests for the one-sample location problem

This paper studies the effect of the normal distribution assumption on the power and size of the sign test, Wilcoxon's signed rank test and the t-test when used in one-sample location problems. Power functions for these tests under various skewness and kurtosis conditions are produced for several sample sizes from simulated data using the g-and-k distribution of MacGillivray and Cannon.

Latent structure of mentoring function scales

The latent structure of a 16-item mentoring function instrument was analyzed with the responses of 568 full-time employees. Features of the analyses included (a) assessment of the items' distributional properties (i.e., skewness and kurtosis); (b) assessment of the factor structure using the Satorra-Bentler scaled test statistic; and (c) evaluation of the instrument's invariance across sex. Confirmatory factor analyses using the scaled chi-square supported a two-factor oblique model that consisted of psychosocial and career-related mentoring functions. The invariance tests suggested that the structure was invariant across sex groups.

Selection of the right risk measures for portfolio allocation

 An optimisation framework is proposed to enable investors to select the right risk measures in portfolio selection. Verification is deployed by performing experiments in developed markets (e.g., the US stock market), emerging markets (e.g., the South Korean stock market) and global investments. A preselection procedure dealing with large datasets is also introduced to eliminate stocks that have low diversification potential before running the portfolio optimisation model. Portfolios are evaluated by four performance indices, i.e., the Sortino ratio, the Sharpe ratio, the Stutzer performance index, and the Omega measure. Experimental results demonstrate that high performance and also well-diversified portfolios are obtained if modified value-at-risk, variance, or semi-variance is concerned whereas emphasising only skewness, kurtosis or higher moments in general produces low performance and poorly diversified portfolios. In addition, the preselection applied to large datasets results in portfolios that have not only high performance but also high diversification degree.

What determines the yen swap spread?

We investigate if Japanese yen denominated interest rate swap spreads price risks in addition to liquidity and default risk. These additional risks include: the time-varying correlation between interest rates of different types and maturities; business cycle risk; and market skewness risk. Our analysis, over a number of different maturities and sample periods, supports the existence of an additional risk premium. We also show that the time-varying correlation between short term market interest rates (e.g., TIBOR) and the longer term Government bond yield (e.g., Gensaki) is of particular importance. Japanese yen swap spreads are shown to contain both pro-cyclical and counter-cyclical elements of business cycle risk, positive risk premia for skewness risk and variable risk premia for correlation risk (between fixed and floating interest rates).

Bacterial attachment on sub-nanometrically smooth titanium substrata

Despite the volume of work that has been conducted on the topic, the role of surface topography in mediating bacterial cell adhesion is not well understood. The primary reason for this lack of understanding is the relatively limited extent of topographical characterisation employed in many studies. In the present study, the topographies of three sub-nanometrically smooth titanium (Ti) surfaces were comprehensively characterised, using nine individual parameters that together describe the height, shape and distribution of their surface features. This topographical analysis was then correlated with the adhesion behaviour of the pathogenic bacteria Staphylococcus aureus and Pseudomonas aeruginosa, in an effort to understand the role played by each aspect of surface architecture in influencing bacterial attachment. While P. aeruginosa was largely unable to adhere to any of the three sub-nanometrically smooth Ti surfaces, the extent of S. aureus cell attachment was found to be greater on surfaces with higher average, RMS and maximum roughness and higher surface areas. The cells also attached in greater numbers to surfaces that had shorter autocorrelation lengths and skewness values that approached zero, indicating a preference for less ordered surfaces with peak heights and valley depths evenly distributed around the mean plane. Across the sub-nanometrically smooth range of surfaces tested, it was shown that S. aureus more easily attached to surfaces with larger features that were evenly distributed between peaks and valleys, with higher levels of randomness. This study demonstrated that the traditionally employed amplitudinal roughness parameters are not the only determinants of bacterial adhesion, and that spatial parameters can also be used to predict the extent of attachment.

Portfolio selection under higher moments using fuzzy multi-objective linear programming

Since asset returns have been recognized as not normally distributed, the avenue of research regarding portfolio higher moments soon emerged. To account for uncertainty and vagueness of portfolio returns as well as of higher moment risks, we proposed a new portfolio selection model employing fuzzy sets in this paper. A fuzzy multi-objective linear programming (MOLP) for portfolio optimization is formulated using marginal impacts of assets on portfolio higher moments, which are modelled by trapezoidal fuzzy numbers. Through a consistent centroid-based ranking of fuzzy numbers, the fuzzy MOLP is transformed into an MOLP that is then solved by the maximin method. By taking portfolio higher moments into account, the approach enables investors to optimize not only the normal risk (variance) but also the asymmetric risk (skewness) and the risk of fat-tails (kurtosis). An illustrative example demonstrates the efficiency of the proposed methodology comparing to previous portfolio optimization models.

A transformation technique to estimate the process capability index for non-normal processes

Estimating the process capability index (PCI) for non-normal processes has been discussed by many researches. There are two basic approaches to estimating the PCI for non-normal processes. The first commonly used approach is to transform the non-normal data into normal data using transformation techniques and then use a conventional normal method to estimate the PCI for transformed data. This is a straightforward approach and is easy to deploy. The alternate approach is to use non-normal percentiles to calculate the PCI. The latter approach is not easy to implement and a deviation in estimating the distribution of the process may affect the efficacy of the estimated PCI. The aim of this paper is to estimate the PCI for non-normal processes using a transformation technique called root transformation. The efficacy of the proposed technique is assessed by conducting a simulation study using gamma, Weibull, and beta distributions. The root transformation technique is used to estimate the PCI for each set of simulated data. These results are then compared with the PCI obtained using exact percentiles and the Box-Cox method. Finally, a case study based on real-world data is presented.