The identification of peaks or maxima in probability densities, by mode testing or bump hunting, has become an important problem in applied fields. This task has been approached in the statistical literature from different perspectives, with the proposal of testing procedures which are based on kernel density estimators or on the quantification of excess mass. However, none of the existing proposals provides a satisfactory performance in practice. In this work, a new procedure which combines the previous approaches (smoothing and excess mass) is presented and compared with the existing methods, showing a superior behaviour. A real data example on philatelic data is also included for illustration purposes.

No abstract

A new plug-in rule procedure for bandwidth selection in kernel circular density estimation is introduced. The performance of this proposal is checked throughout a simulation study considering a variety of circular distributions exhibiting multimodality, peakedness and/or skewness. The plug-in rule behaviour is also compared with other existing bandwidth selectors. The method is illustrated with two classical datasets of cross-beds layers and animal orientation.

Postural control is achieved through the integration at the central nervous system level of information obtained by the visual, somatosensory and vestibular systems. Computerized dynamic posturography and the Sway Star system are both used to carry out sensory analysis. The purpose of this study was to determine the influence of sex and age on sensory analysis, measured with these two systems, and to compare their results. A prospective trial was conducted with 70 healthy individuals (average age: 44.9 years) uniformly distributed in seven age groups, who underwent postural study with both systems. We used SPSS 16.0 for statistical study: comparison of means test for influence of gender and age and Pearson's correlation test (p < 0.05). Gender variable had no influence. The influence of age in vestibular input was found to be significant with both posturography systems, while visual input was only found to be significant with the Sway Star. The results with the two systems were not comparable. Sensory contribution does not remain stable throughout life. Visual information decreases with age, reaching a minimum at 40-49 years, and may correspond to the deterioration of eyesight with age. Propioceptive information showed no statistically significant changes, and several forms of treatment might correct the deterioration of this system. Vestibular information reaches a maximum in the 40-49 years age group in an attempt to compensate for visual deterioration, and decreases again in subsequent decades. This may be due to aging of the vestibular system and the difficulty in its correction.

In several applied fields, multimodality assessment is a crucial task as a previous exploratory tool or for determining the suitability of certain distributions. The goal of this paper is to present the utilities of the R package multimode, which collects different exploratory and testing non-parametric approaches for determining the number of modes and their estimated location. Specifically, some graphical tools (SiZer map, mode tree or mode forest) are provided, allowing for the identification of mode patterns, based on the kernel density estimation. Several formal testing procedures for determining the number of modes are described in this paper and implemented in the multimode package, including methods based on the ideas of the critical bandwidth, the excess mass or using a combination of both. This package also includes a function for estimating the modes locations and different classical data examples that have been considered in mode testing literature.