Peter K. Dunn

Series evaluation of Tweedie exponential dispersion model densities

Exponential dispersion models, which are linear exponential families with a dispersion parameter, are the prototype response distributions for generalized linear models. The Tweedie family comprises those exponential dispersion models with power mean-variance relationships. The normal, Poisson, gamma and inverse Gaussian distributions belong to theTweedie family. Apart from these special cases, Tweedie distributions do not have density functions which can be written in closed form. Instead, the densities can be represented as infinite summations derived from series expansions. This article describes how the series expansions can be summed in an numerically efficient fashion. The usefulness of the approach is demonstrated, but full machine accuracy is shown not to be obtainable using the series expansion method for all parameter values. Derivatives of the density with respect to the dispersion parameter are also derived to facilitate maximum likelihood estimation. The methods are demonstrated on two data examples and compared with with Box-Cox transformations and extended quasi-likelihoood.

Does a rater's familiarity with a candidate's pronunciation affect the rating in oral proficiency interviews?

This study investigated factors that could affect inter-examiner reliability in the pronunciation assessment component of speaking tests. We hypothesized that the rating of pronunciation is susceptible to variation in assessment due to the amount of exposure examiners have to nonnative English accents. An inter-rater variability analysis was conducted on the English pronunciation ratings of three test candidate interlanguages: Chinese, Korean, and Indian English. Pronunciation was rated by 99 International English Language Testing System (IELTS) examiners across five geographically dispersed test centres. The examiners had either prolonged exposure, or no, or little exposure to the interlanguage of the candidates. A significant proportion of examiners rated pronunciation higher when they had prolonged exposure, and lower when they had no, or little, exposure to the candidates’ interlanguage. The location of the test centre also had a significant effect on the pronunciation rating, independent of the familiarity variable, with a significant proportion of NNS raters scoring candidates from their home country higher than those who were not. It is recommended that interlanguage phonology familiarity should be considered in the design of speaking tests and rater training and that research is required into test centre bias and the phonological judgements and awareness of OPI raters.

Evaluation of Tweedie exponential dispersion model densities by Fourier inversion

Abstract The Tweedie family of distributions is a family of exponential dispersion models with power variance functions V(μ)=μp for

Estimating crop area using seasonal time series of Enhanced Vegetation Index from MODIS satellite imagery

p\not\in(0,1)

Mobile-phone-based classroom response systems: Students’ perceptions of engagement and learning in a large undergraduate course

. These distributions do not generally have density functions that can be written in closed form. However, they have simple moment generating functions, so the densities can be evaluated numerically by Fourier inversion of the characteristic functions. This paper develops numerical methods to make this inversion fast and accurate. Acceleration techniques are used to handle oscillating integrands. A range of analytic results are used to ensure convergent computations and to reduce the complexity of the parameter space. The Fourier inversion method is compared to a series evaluation method and the two methods are found to be complementary in that they perform well in different regions of the parameter space.

Comparisons of Tobit, Linear, and Poisson-Gamma Regression Models An Application of Time Use Data

Cereal grain is one of the main export commodities of Australian agriculture. Over the past decade, crop yield forecasts for wheat and sorghum have shown appreciable utility for industry planning at shire, state, and national scales. There is now an increasing drive from industry for more accurate and cost-effective crop production forecasts. In order to generate production estimates, accurate crop area estimates are needed by the end of the cropping season. Multivariate methods for analysing remotely sensed Enhanced Vegetation Index (EVI) from 16-day Moderate Resolution Imaging Spectroradiometer (MODIS) satellite imagery within the cropping period (i.e. April-November) were investigated to estimate crop area for wheat, barley, chickpea, and total winter cropped area for a case study region in NE Australia. Each pixel classification method was trained on ground truth data collected from the study region. Three approaches to pixel classification were examined: (i) cluster analysis of trajectories of EVI values from consecutive multi-date imagery during the crop growth period; (ii) harmonic analysis of the time series (HANTS) of the EVI values; and (iii) principal component analysis (PCA) of the time series of EVI values. Images classified using these three approaches were compared with each other, and with a classification based on the single MODIS image taken at peak EVI. Imagery for the 2003 and 2004 seasons was used to assess the ability of the methods to determine wheat, barley, chickpea, and total cropped area estimates. The accuracy at pixel scale was determined by the percent correct classification metric by contrasting all pixel scale samples with independent pixel observations. At a shire level, aggregated total crop area estimates were compared with surveyed estimates. All multi-temporal methods showed significant overall capability to estimate total winter crop area. There was high accuracy at pixel scale (>98% correct classification) for identifying overall winter cropping. However, discrimination among crops was less accurate. Although the use of single-date EVI data produced high accuracy for estimates of wheat area at shire scale, the result contradicted the poor pixel-scale accuracy associated with this approach, due to fortuitous compensating errors. Further studies are needed to extrapolate the multi-temporal approaches to other geographical areas and to improve the lead time for deriving cropped-area estimates before harvest.

Walk Score and Australian adults' home-based walking for transport

Using a Classroom Response System (CRS) has been associated with positive educational outcomes, by fostering student engagement and by allowing immediate feedback to both students and instructors. This study examined a low-cost CRS (VotApedia) in a large first-year class, where students responded to questions using their mobile phones. This study explored whether the use of VotApedia retained the advantages of other CRS, overcame some of the challenges of other CRS, and whether new challenges were introduced by using VotApedia. These issues were studied within three themes: students’ perceptions of using VotApedia; the impact of VotApedia on their engagement; and the impact of VotApedia on their learning. Data were collected from an online survey, focus groups and student feedback on teaching and course content. The results indicated that using VotApedia retains the pedagogical advantages of other CRS, while overcoming some of the challenges presented by using other CRS, without introducing any new challenges.

The use of a Classroom Response System to more effectively flip the classroom

Time use data (TUD) are distinctive, being episodic in nature and consisting of both continuous and discrete (exact zeros) values. TUD is non-negative and generally right skewed. To analyze such data, the Tobit, and to a lesser extent, linear regression models are often used. Tobit models assume the zeros represent censored values of an underlying normally distributed latent variable that theoretically includes negative values. Both the linear regression and Tobit models have normality as a key assumption. The Poisson-gamma distribution is a distribution with both a point mass at zero (corresponding to zero time spent on a given activity) and a continuous component. Using generalized linear models, TUD can be modeled utilizing the Poisson-gamma distribution. Using TUD, Tobit and linear regression models are compared to the Poisson-gamma with respect to the interpretation of the model, the model fit (analysis of residuals), and model performance through the use of a simulated data experiment. The Poisson-gamma is found to be theoretically and empirically more sound in many circumstances.

Instructor Perceptions of Using a Mobile-Phone-Based Free Classroom Response System in First-Year Statistics Undergraduate Courses

The relationships of Walk Score, a publicly-accessible walkability assessment tool, with walking for transport to and from home were examined among a large representative sample of Australian adults aged 18-64 years (N=16,944). Residents in highly and somewhat walkable areas were twice and 1.4 times more likely to accumulate 30 min of walking per day compared to those in very car-dependent neighborhoods, respectively. Mean duration of walking was also longer for participants living in highly and somewhat walkable areas compared to those in very car-dependent areas. Walk Score has potential as a widely-applicable tool for identifying the walkability of local neighborhoods.

Estimating winter crop area across seasons and regions using time-sequential MODIS imagery

Classroom Response Systems (CRS) have been shown to improve student learning outcomes by encouraging student engagement with the course content, instructors and student peers. The systems ability to provide immediate and quality feedback to both students and instructors, particularly in large classes, is highly desirable. While CRS has been used for well over a decade and been shown to successfully improve student engagement and participation, a number of studies have also identified that its use could potentially mean that less material is able to be covered in lectures. Clearly, the approach of cramming CRS into already content-heavy class time does not embrace the potential for CRS to improve student engagement and student learning. The use of CRS should be planned as an integral component of the course which enhances and reinforces the learning outcomes. The effectiveness of CRS depends strongly on the quality and variety of the questions, and the design of the activities to encourage students to engage with the questions. This case study explores the use of a new, low-cost, state-ofthe-art CRS (Top Hat Monocle) which allows students to use their mobile devices (phones, tablets, laptops) to respond to a variety of numerical, multiple-choice, short-answer and open-ended discussion questions posed during face-to-face workshops. In order to allow sufficient time to fully engage with the workshop activities traditional lectures were revised and the classroom lecture was flipped. Students worked through narrated lecture material (hand-e-lectures) online, prior to attending the workshops. CRS was included as part of the e-lecture content and feedback from this was incorporated into the workshops. Workshops extended the e-lecture content by including a variety of carefully designed, engaging activities (many were group activities) that used CRS questions to facilitate discussions, problem solving and case study analysis to enhance student cognition. Overall, the new flipped lecture and CRS teaching format demonstrated a substantial increase in the level of student engagement, motivation and attendance compared to previous cohorts.