Search

The teaching and learning of statistical thinking begins at a young age in Australia, with a focus on data representation and interpretation from Foundation Year (age 5), and the collection, sorting and categorising of items from the natural environment starting even earlier. The intangible concept of data, as part of statistical literacy, can be complex for children to grasp, especially when applying the notion of data to the everyday world or when data are explored in isolation to an investigation process. Authentic data modelling experiences present meaningful opportunities to apply statistical thinking although expert STEM knowledge is not always accessible to primary classroom teachers, nor is it always obvious how to implement such authentic problems within a classroom context. In this exploratory case study, we present data from a Year 4 classroom (age 9) statistical investigation addressing, 'How big is a leaf?' linking data to the real-life STEM context they represented. The authors were interested in how the teacher's communication processes supported her students' emerging understandings about data. Wit's (2018) cognitive tuning framework offered a way to capture how the communication processes in a group build to a commonly shared frame of reference. Findings revealed a pattern of communication between the teacher and students, supporting students' changing conceptions of data and related statistical thinking processes, throughout the investigation. [ABSTRACT FROM AUTHOR]

Late talking children are at risk of ongoing language impairment. This intervention study replicated and extended research based on cross-situational statistical learning principles. Three late talking children (age 24–32 months) were enrolled into the concurrent multiple baseline single-case experimental intervention study. The intervention consisted of 16 sessions over eight/nine weeks, including 10–11 pairs of target and control words (three per session). Children heard the target words a minimum of 64 times per session, in sentences with high linguistic variability in varied play activities. All children increased production of target words and expressive vocabulary, with statistically significant differences between word acquisition in baseline and intervention phases. One of the three children learnt statistically significantly more target words than control words. The results replicated the findings of previous research for some but not all of the participants, providing individual evidence that this approach has promise as a therapy technique for late talking children. [ABSTRACT FROM AUTHOR]

Background: Kidney graft failure risk prediction models assist evidence-based medical decision-making in clinical practice. Our objective was to develop and validate statistical and machine learning predictive models to predict death-censored graft failure following deceased donor kidney transplant, using time-to-event (survival) data in a large national dataset from Australia.Methods: Data included donor and recipient characteristics (n = 98) of 7,365 deceased donor transplants from January 1st, 2007 to December 31st, 2017 conducted in Australia. Seven variable selection methods were used to identify the most important independent variables included in the model. Predictive models were developed using: survival tree, random survival forest, survival support vector machine and Cox proportional regression. The models were trained using 70% of the data and validated using the rest of the data (30%). The model with best discriminatory power, assessed using concordance index (C-index) was chosen as the best model.Results: Two models, developed using cox regression and random survival forest, had the highest C-index (0.67) in discriminating death-censored graft failure. The best fitting Cox model used seven independent variables and showed moderate level of prediction accuracy (calibration).Conclusion: This index displays sufficient robustness to be used in pre-transplant decision making and may perform better than currently available tools. [ABSTRACT FROM AUTHOR]

This paper develops an integrated machine learning and enhanced statistical approach for wind power interval forecasting. A time-series wind power forecasting model is formulated as the theoretical basis of our method. The proposed model takes into account two important characteristics of wind speed: the nonlinearity and the time-changing distribution. Based on the proposed model, six machine learning regression algorithms are employed to forecast the prediction interval of the wind power output. The six methods are tested using real wind speed data collected at a wind station in Australia. For wind speed forecasting, the long short-term memory (LSTM) network algorithm outperforms other five algorithms. In terms of the prediction interval, the five nonlinear algorithms show superior performances. The case studies demonstrate that combined with an appropriate nonlinear machine learning regression algorithm, the proposed methodology is effective in wind power interval forecasting. [ABSTRACT FROM AUTHOR]

The climate system is an excellent example of a "complex system" since there is an interplay and inter-relation of several climate variables. Variables such as the Standardized Precipitation Index (SPI) are used to indicate the drought climate condition – a negative (or positive) value of SPI would imply a dry (or wet) state in a region. It is difficult to identify the factors influencing the SPI or their inter-relations (including feedback loops). Here, we aim to study the complex dynamics that SPI has with the sea surface temperature (SST), El Niño Southern Oscillation (ENSO) (aka., NINO 3.4) and Indian Ocean Dipole (IOD), using a machine learning approach. Our findings are: (i) IOD was negatively correlated to SPI till 2008; (ii) until 2004, SST was negatively correlated with SPI; (iii) from 2005 to 2014, the SST had swung between negative and positive correlations; (iv) since 2014, we observed that the regression coefficient (δ) corresponding to SST has always been positive; (v) the SST has an upward trend, and the positive upward trend of δ implied that SPI has been positively correlated with SST in recent years; and finally, (vi) the current value of SPI has a significant positive correlation with a past SPI value with a periodicity of about 7.5 years. Examining the complex dynamics, we used a statistical machine learning approach to construct an inferential network of these climate variables, which revealed that SST and NINO 3.4 directly couples with SPI, whereas IOD indirectly couples with SPI through SST and NINO 3.4. The system also indicated that Nino 3.4 significantly negatively affects SPI. If Nino 3.4 increases (decreases), the SPI drops (increases) considerably, leading to more drought (wet) conditions. Interestingly, there was a structural change in the complex dynamics of the four climate variables of NINO 3.4, IOD, SST, and SPI, sometime in 2008. Though a simple 12-month moving average of SPI over 58 years (1961–2018) has a negative trend towards drought, the complex dynamics of SPI with other climate variables indicate a wet season for south-west Australia. [ABSTRACT FROM AUTHOR]

Local and state governments depend on small area population forecasts to make important decisions concerning the development of local infrastructure and services. Despite their importance, current methods often produce highly inaccurate forecasts. Recent years have witnessed promising developments in time series forecasting using Machine Learning across a wide range of social and economic variables. However, limited work has been undertaken to investigate the potential application of Machine Learning methods in demography, particularly for small area population forecasting. In this paper we describe the development of two Long-Short Term Memory network architectures for small area populations. We employ the Keras Tuner to select layer unit numbers, vary the window width of input data, and apply a double training and validation regime which supports work with short time series and prioritises later sequence values for forecasts. These methods are transferable and can be applied to other data sets. Retrospective small area population forecasts for Australia were created for the periods 2006–16 and 2011–16. Model performance was evaluated against actual data and two benchmark methods (LIN/EXP and CSP-VSG). We also evaluated the impact of constraining small area population forecasts to an independent national forecast. Forecast accuracy was influenced by jump-off year, constraining, area size, and remoteness. The LIN/EXP model was the best performing method for the 2011-based forecasts whilst deep learning methods performed best for the 2006-based forecasts, including significant improvements in the accuracy of 10 year forecasts. However, benchmark methods were consistently more accurate for more remote areas and for those with populations <5000. • A novel deep learning method for small area forecasts is presented. • Forecast accuracy was influenced by jump-off year, constraining, area size, and remoteness. • Deep learning methods performed best for the 5 and 10 year 2006-based forecasts. • The LIN/EXP model was the best performing method for the 5 year 2011-based forecasts. • Benchmark methods were consistently more accurate for more remote areas and for those with populations <5000. [ABSTRACT FROM AUTHOR]

Property valuation plays a significant role in urban economics and is of great importance to various stakeholders who interact and shape the city, including property owners, buyers, banks, land developers, real estate agents, local councils and government planning authorities. In the literature, various predictive models have been proposed to automate the calculation of property value, most of which endeavour to factor in the combination of property characteristics, market factors and location-based attributes associated with individual properties use large citywide databases. At the same time, it has been widely acknowledged that regional sub-areas have impacts on property price prediction. Therefore, this paper aims to investigate the performance of various techniques on sub-areas using the Greater Sydney Region as the study area. The sub-area in this paper is defined as the statistical areas (SAs) as defined by the Australian Bureau of Statistics. In particular, two different SA geographies (SA4, SA3) along with the City Level are adopted to understand the spatial dependence which occurs at different levels. With real-world transaction records and data collected from a diverse range of sources, various methods including the traditional hedonic price model (HPM) and popular machine learning (ML) approaches are implemented and evaluated for property price prediction. Two different property markets for residential property are modelled, being for housing stock and apartment (unit) stock. Experimental results show that Random Forest and Gradient Boosting-based methods outperform other approaches in most scenarios and that the high spatial resolution property sub-area (SA3) improved the performance in terms of overall model accuracy. This research provides insights into how sub-area machine learning models can be employed in real estate to characterize property price, and helps understand the influential factors in different local geographical areas for policy-making. • Property valuation is modelled with property attributes, locational and socioeconomic factors by a variety of machine learning algorithms. • Models are implemented on three different geographical area levels. • Both in-time and out-of-time validation are implemented. • House and Apartments are both valued. • Valuation carried out on smaller regions can improve the predictive performance of most modelling techniques. • The study illustrates that the variables' contributions to the property price vary for different geographical areas. [ABSTRACT FROM AUTHOR]

Background: Early or first-line treatment with biologics, as opposed to conventional immunomodulators, is not always necessary to achieve remission in Crohn's disease [CD] and may not be cost-effective. This study aimed to develop a simple model to predict the need for early biologic therapy, in order to risk-stratify CD patients and guide initial treatment selection., Methods: A model-building study using supervised statistical learning methods was conducted using a retrospective cohort across two tertiary centres. All biologic-naïve CD patients who commenced an immunomodulator between January 1, 2004 and December 31, 2016, were included. A predictive score was derived using Cox regression modelling of immunomodulator failure, and was internally validated using bootstrap resampling., Results: Of 410 patients [median age 37 years, 47% male, median disease duration 4.7 years], 229 [56%] experienced immunomodulator failure [39 required surgery, 24 experienced a new stricture, 44 experienced a new fistula/abscess, 122 required biologic escalation] with a median time to failure of 16 months. Independent predictors of treatment failure included raised C-reactive protein [CRP], low albumin, complex disease behaviour, younger age, and baseline steroids. Highest CRP and lowest albumin measured within the 3 months preceding immunomodulator initiation outperformed baseline measurements. After model selection, only highest CRP and lowest albumin remained and the resultant Crohn's Immunomodulator CRP-Albumin [CICA] index demonstrated robust optimism-corrected discriminative performance at 12, 24, and 36 months (area under the curve [AUC] 0.84, 0.83, 0.81, respectively)., Conclusions: The derived CICA index based on simple, widely available markers is feasible, internally valid, and has a high utility in predicting immunomodulator failure. This requires external, prospective validation., (© The Author(s) 2020. Published by Oxford University Press on behalf of European Crohn’s and Colitis Organisation. All rights reserved. For permissions, please email: journals.permissions@oup.com.)

Objectives: To (1) use an elastic net (EN) algorithm to derive a frailty measure from a national aged care eligibility assessment program; (2) compare the ability of EN-based and a traditional cumulative deficit (CD) based frailty measures to predict mortality and entry into permanent residential care; (3) assess if the predictive ability can be improved by using weighted frailty measures., Materials and Methods: A Cox proportional hazard model based EN algorithm was applied to the 2003-2013 cohort of 903 996 participants for selecting items to enter an EN based frailty measure. The out-of-sample predictive accuracy was measured by the area under the curve (AUC) from Cox models fitted to 80% training and validated on 20% testing samples., Results: The EN approach resulted in a 178-item frailty measure including items excluded from the 44-item CD-based measure. The EN based measure was not statistically significantly different from the CD-based approach in terms of predicting mortality (AUC 0.641, 95% CI: 0.637-0.644 vs AUC 0.637, 95% CI: 0.634-0.641) and permanent care entry (AUC 0.626, 95% CI: 0.624-0.629 vs AUC 0.627, 95% CI: 0.625-0.63). However, the weighted EN based measure statistically outperforms the weighted CD measure for predicting mortality (AUC 0.774, 95% CI: 0.771-0.777 vs AUC 0.757, 95% CI: 0.754-0.760) and permanent care entry (AUC 0.676, 95% CI: 0.673-0.678 vs AUC 0.671, 95% CI: 0.668-0.674)., Conclusions: The weighted EN and CD-based measures demonstrated similar prediction performance. The CD-based measure items are relevant to frailty measurement and easier to interpret. We recommend using the weighted and unweighted CD-based frailty measures., (© The Author(s) 2020. Published by Oxford University Press on behalf of the American Medical Informatics Association. All rights reserved. For permissions, please email: journals.permissions@oup.com.)