Bayesian regression filter and the issue of priors

We propose a Bayesian framework for regression problems, which covers areas which are usually dealt with by function approximation. An online learning algorithm is derived which solves regression problems with a Kalman filter. Its solution always improves with increasing model complexity, without the risk of over-fitting. In the infinite dimension limit it approaches the true Bayesian posterior. The issues of prior selection and over-fitting are also discussed, showing that some of the commonly held beliefs are misleading. The practical implementation is summarised. Simulations using 13 popular publicly available data sets are used to demonstrate the method and highlight important issues concerning the choice of priors.

School outcomes:sharing the responsibility between pupil and school

This paper uses a Data Envelopment Analysis based approach to decompose pupil under-attainment into that attributable to the school the pupil attends and that attributable to the pupil. The approach measures pupil attainment in terms of value added. Data on over 6700 A-level pupils from 122 English schools have been analysed. The results suggest that at current levels of school effectiveness a pupil's own application accounts for the major part of any under-attainment, though schools also have scope to improve their effectiveness.The approach also makes it possible to identify target attainment levels a pupil could be set and the extent to which the attainment of those targets necessitates an improvement in the effectiveness of the school the pupil attends and in the pupil's own efforts.

Assessing pupil and school performance by non-parametric and parametric techniques

This paper discusses the use of the non-parametric free disposal hull (FDH) and the parametric multi-level model (MLM) as alternative methods for measuring pupil and school attainment where hierarchical structured data are available. Using robust FDH estimates, we show how to decompose the overall inefficiency of a unit (a pupil) into a unit specific and a higher level (a school) component. By a sample of entry and exit attainments of 3017 girls in British ordinary single sex schools, we test the robustness of the non-parametric and parametric estimates. Finally, the paper uses the traditional MLM model in a best practice framework so that pupil and school efficiencies can be computed.

Simultaneous interrogation of fiber Bragg grating sensors using an acoustooptic tunable filter

We describe a novel technique to provide demultiplexing of fiber Bragg grating sensors, interrogated using interferometric wavelength shift detection. Amplitude modulation of multiple radio frequency driving signals allows an acoustooptic tunable filter to provide wavelength demultiplexing. We demonstrated a noise limited strain resolution of 150 nanostrain/v(Hz) and a crosstalk better than -50 dB.

Fast and widely tunable Bragg grating reflection filter

A fiber Bragg grating filter device linearly tunable over 45 nm is presented. The device has a maximum tuning speed of 19 nm/ms with a wavelength setting time below 1.5 ms.

Portfolio diversification and filter rule profits

Previously, it has been shown that the profits from a simple market timing trading rule applied to a portfolio of shares can be affected by the inter-relationships between the returns of the component securities. In this short letter, the results from applying a more sophisticated 'filter' rule to the same data are reported. Unlike the simple trading rule, the filter rule does produce some evidence of economic profits.

Assessing pupil and school performance by non-parametric and parametric techniques

Often observations are nested within other units. This is particularly the case in the educational sector where school performance in terms of value added is the result of school contribution as well as pupil academic ability and other features relating to the pupil. Traditionally, the literature uses parametric (i.e. it assumes a priori a particular function on the production process) Multi-Level Models to estimate the performance of nested entities. This paper discusses the use of the non-parametric (i.e. without a priori assumptions on the production process) Free Disposal Hull model as an alternative approach. While taking into account contextual characteristics as well as atypical observations, we show how to decompose non-parametrically the overall inefficiency of a pupil into a unit specific and a higher level (i.e. a school) component. By a sample of entry and exit attainments of 3017 girls in British ordinary single sex schools, we test the robustness of the non-parametric and parametric estimates. We find that the two methods agree in the relative measures of the scope for potential attainment improvement. Further, the two methods agree on the variation in pupil attainment and the proportion attributable to pupil and school level.

Pulse tail suppression in laser diode output by tunable notch filter

Efficient suppression of relaxation oscillations in the output signal from an overdriven gain-switched laser diode was demonstrated. Several quantum-well distributed feedback laser diodes from different manufacturers were used for experimental analysis. A five-fold increase in the peak power was achieved for the tail-free operation. It was found that spectral filtering removed the nonlinearly chirped components resulting in pulse shortening by a factor of three.

Bidirectional nonreciprocal wavelength-interleaving coherent fiber transversal filter

A bidirectional nonreciprocal wavelength-interleaving filter based on an optically coherent high birefringence fiber transversal filter structure is demonstrated. Stable, low loss, low dispersion, and high isolation operation is demonstrated with reconfigurable transfer characteristics for interleaved channel spacing of 0.8 nm.

Bragg Grating fast Tunable Filter For Wavelength Division Multiplexing

A Bragg grating fast tunable filter prototype working over a linear tuning range of 45 nm with a maximum tuning speed of 21 nm/ms has been realized. The tunable filter system is based on two piezoelectric stack actuators moving a mechanical device thus compressing an apodized fiber Bragg grating. The filter allows both traction and compression and can work in transmission and in reflection. It is designed to work with a channel spacing of 100 GHz according to the ITU specifications for wavelength division multiplexing systems

The ecology of percolating filters containing a plastic filter medium in relation to their efficiency in The treatment of domestic sewage

The suitability of a new plastic supporting medium for biofiltration was tested over a three year period. Tests were carried out on the stability, surface properties, mechanical strength, and dimensions of the medium. There was no evidence to suggest that the medium was deficient in any of these respects. The specific surface (320m2m-3) and the voidage (94%) of the new medium are unlike any other used in bio-filtration and a pilot plant containing two filters was built to observe its effects on ecology and performance. Performance was estimated by chemical analysis and ecology studied by film examination and fauna counts. A system of removable sampling baskets was designed to enable samples to be obtained from two intermediate depths of filter. One of the major operating problems of percolating filters is excessive accumulation of film. The amount of film is influenced by hydraulic and organic load and each filter was run at a different loading. One was operated at 1.2m3m-3day-1 (DOD load 0.24kgm-3day-1) judged at the time to be the lowest filtration rate to offer advantages over conventional media. The other filter was operated at more than twice this loading (2.4m3m-3day-lBOD load 0.55kgm-3day-1) giving a roughly 2.5x and 6x the conventional loadings recommended for a Royal Commission effluent. The amount of film in each filter was normally low (0.05-3kgm(3 as volatile solids) and did not affect efficiency. The evidence collected during the study indicated that the ecology of the filters was normal when compared with the data obtained from the literature relating to filters with mineral media. There were indications that full ecological stability was yet to be reached and this was affecting the efficiency of the filters. The lower rate filter produced an average 87% BOD removal giving a consistent Royal Commission effluent during the summer months. The higher rate filter produced a mean 83% BOD removal but at no stage a consistent Royal Commission effluent. From the data on ecology and performance the filters resembled conventional filters rather than high rate filters.

Pupil resistance to their schooling experience

The thesis explores the nature of pupil resistance; it investigates what constitutes it and how it can be explained. Various ethnic and national group, male and female working-class resistance if analysed in two secondary schools in Birmingham (England) and one school in Sydney (Australia). It focuses on the pupils’ experience of school. ‘Compressed ethnographies’ (Walford and Miller, 1991) were conducted in each school to examine pupil resistance. The research found that structural societal state factors, regional, community and formal, informal and physical characteristics of each school, together with the teachers and pupils characteristics and background all influence resistance. The class, gender, ethnic and national identity of each pupil shapes resistance. In all three schools that were involved with the research, girls were more likely to exhibit overt, collective forms of resistance, whereas lads were more likely to operate alone. Islander pupils in Sydney and African-Caribbean kids in Birmingham were more likely to display engaged forms of resistance. Girls tended to show more engaged forms compared to their male counterparts across all ethnic and national cultures. Resistance is complex and dynamic, the definition alters depending upon context. Dimensions of resistance are developed; including overt, covert; individual, collective; intentional, unintentional; engaged and detached forms. Resistance operates within a structure and agency framework, the pupils can shape their own schooling experience mediated within the structures of their school, community and society. Some pupils manage their resources and the structures better than others; how the pupil manages and operates within the structures influences their resistance response. Resistance is contradictory and can reinforce the status quo. To fully understand resistance, it must be contextualised.