102 research outputs found
Short-term real-time prediction of total number of reported COVID-19 cases and deaths in South Africa: a data driven approach
No full textBackground: The rising burden of the ongoing COVID-19 epidemic in South Africa has motivated the application of modeling strategies to predict the COVID-19 cases and deaths. Reliable and accurate short and long-term forecasts of COVID-19 cases and deaths, both at the national and provincial level, are a key aspect of the strategy to handle the COVID-19 epidemic in the country. Methods: In this paper we apply the previously validated approach of phenomenological models, fitting several non-linear growth curves (Richards, 3 and 4 parameter logistic, Weibull and Gompertz), to produce short term forecasts of COVID-19 cases and deaths at the national level as well as the provincial level. Using publicly available daily reported cumulative case and death data up until 22 June 2020, we report 5, 10, 15, 20, 25 and 30-day ahead forecasts of cumulative cases and deaths. All predictions are compared to the actual observed values in the forecasting period. Results: We observed that all models for cases provided accurate and similar short-term forecasts for a period of 5 days ahead at the national level, and that the three and four parameter logistic growth models provided more accurate forecasts than that obtained from the Richards model 10 days ahead. However, beyond 10 days all models underestimated the cumulative cases. Our forecasts across the models predict an additional 23,551-26,702 cases in 5 days and an additional 47,449-57,358 cases in 10 days. While the three parameter logistic growth model provided the most accurate forecasts of cumulative deaths within the 10 day period, the Gompertz model was able to better capture the changes in cumulative deaths beyond this period. Our forecasts across the models predict an additional 145-437 COVID-19 deaths in 5 days and an additional 243-947 deaths in 10 days. Conclusions: By comparing both the predictions of deaths and cases to the observed data in the forecasting period, we found that this modeling approach provides reliable and accurate forecasts for a maximum period of 10 days ahead.Reddy, T (corresponding author), South African Med Res Council, Biostat Res Unit, Cape Town, South Africa ; Hasselt Univ, Censtat, Hasselt, Belgium.
[email protected]
New characterisations of spatial linear networks for geographical accessibility
No full textThesis (PhD (Mathematical Statistics))--University of Pretoria, 2024.Target 9.1 of the United Nations Sustainable Development Goals specifies the need for affordable, equitable access for all. In South Africa, where most travel occurs via the road network, apartheid policies designed the historical road network to segregate rather than integrate. Since the end of apartheid, there has been an increased need for integrated urban accessibility. Since government initiatives are typically enacted at a regional level, it is relevant to model accessibility between regions. Very few methods exist in the literature that model road-based inter-regional accessibility, and none account for structural characteristics of the road network. The aim of this thesis is to develop a novel stochastic model that estimates road-based inter-regional accessibility, and that is able to take the homogeneity of road networks into account. The accessibility model utilises Markov chain theory. Each region represents a state, and the average inverse distances between regions act as transition probabilities. Transition probabilities between adjacent regions are stored in a 1-step transition probability matrix (TPM). Assuming the Markov property holds, raising the TPM to the power n gives transition probabilities between regions up to n steps away. Letting n→∞ gives the prominence index, which quantifies the accessibility of a region regardless of the journey’s starting point. Road network homogeneity is tested by extending a test for the homogeneity of spatial point patterns to spatial linear networks. An unsupervised clustering method is then developed which subdivides a road network into regions that are as homogeneous as possible. Finally, road-based accessibility is calculated between these regions. The accessibility model was first applied to electoral wards in the City of Tshwane. Based on the wards, the central business district (CBD) was most accessible, but there was poor accessibility to the CBD from outlying townships. The homogeneity test showed that distinct residential neighbourhoods were internally homogeneous, and was thus able to identify neighbourhoods within a road network. The unsupervised clustering method was then used to identify two new regionalisations of the road network within the City of Tshwane at different spatial scales, and the accessibility model was applied to these regionalisations. For one regionalisation, an emerging economic area was most accessible, while for the other, a central educational area was most accessible. Although accessibility was not correlated with road network homogeneity, different spatial scales and regionalisations had a great impact on the accessibility results. This thesis develops a new characterisation of spatial linear networks based on their homogeneity, and uses this to investigate the state of inter-regional road-based accessibility in the City of Tshwane. This is a crucial area of research in the move towards a more equitable and sustainable future.National Research Foundation of South Africa under Grant 137785National Research Foundation of South Africa under CoE-MaSS ref #2022-018-MAC-RoadNRF-SASA Academic Statistics BursaryStatisticsPhD (Mathematical Statistics)RestrictedFaculty of Natural and Agricultural Science
Statistical accuracy of an extraction algorithm for linear image objects
No full textMini Dissertation (MSc)--University of Pretoria, 2019.Informal unpaved roads in developing countries arise naturally through human movement and informal housing setups. These roads are not authorised nor maintained by council, nor recorded in official databases or online maps. Mapping such roads from satellite images is a common problem, as information
on these roads is critical for sustainable city growth. Information on their location and extent may be gleaned from spatial big data, however, no automatic or semi-automatic approach is freely available. This research develops a novel algorithm for extracting informal roads from multispectral satellite images, using physical road characteristics. These include near-infrared reflectance, addressed via the NDVI index, shape, addressed via measures of compactness and elongation, and grey-value intensity. The crux of the algorithm is the Discrete Pulse Transform, implemented via the Roadmaker's Pavage. The algorithm provides a classification of road objects, along with an associated uncertainty measure for each road object. Accuracy is assessed using per-pixel assessment metrics and metrics based on road characteristics, including completeness, correctness, and Pratt's Figure of Merit, which is applied to road extraction accuracy for the first time. The algorithm is applied to areas in Gauteng and North West Provinces, South Africa.
Sources of uncertainty and error are discussed, such as indefinite boundaries, surface type heterogeneity, trees and shadows.Acknowledgement of the National Research Foundation for the funding provided through the NRF-SASA Crisis in Academic Statistics grant.StatisticsMScUnrestricte
Distinguishing tree species from in situ hyperspectral and temporal measurements through ensemble statistical learning
No full textThe data presented in this study may be obtained from the corresponding author upon request. Due to intellectual property and confidentiality concerns, the data is not publicly available.Hyperspectral sensors capture and compute spectral reflectance of objects over many
wavelength bands, resulting in a high-dimensional space with enough information to differentiate
between spectrally similar objects. Due to the curse of dimensionality, high spectral dimensionality
can also be difficult to handle and analyse, demanding complex processing and the use of advanced
analytical techniques. Moreover, when hyperspectral measurements are taken at different temporal
frequencies, separation is likely to improve; however, additional complexities in modelling time
variability concurrently with this high spectral dimensionality may be created. As a result, the applicability of ensemble-based techniques suitable for high-dimensional data is examined in this research,
together with the statistical evaluation of time-induced variability, since spectral measurements of tree
species were taken at different time periods. Classification errors for the stochastic gradient boosting
(SGB) and random forest (RF) methods ranged between 5.6% and 13.5%, respectively. Differences in
classification accuracy or errors were also accounted for in the assessment of the models, with up
to 46% of variation in classification error due to the effect of time in the RF model, indicating that
measurement time is important in improving discrimination between tree species. This is because
optical leaf characteristics can vary during the course of the year due to seasonal effects, health status,
or the developmental stage of a tree. Different spectral properties (assumed from relevant wavelength
bands) were found to be key factors impacting the models’ discrimination performance at various
measurement times.The Council for Scientific and Industrial Research (CSIR).https://www.mdpi.com/journal/remotesensingPlant Production and Soil SciencePlant ScienceSDG-15:Life on lan
Distinguishing Tree Species from In Situ Hyperspectral and Temporal Measurements through Ensemble Statistical Learning
No full textHyperspectral sensors capture and compute spectral reflectance of objects over many wavelength bands, resulting in a high-dimensional space with enough information to differentiate between spectrally similar objects. Due to the curse of dimensionality, high spectral dimensionality can also be difficult to handle and analyse, demanding complex processing and the use of advanced analytical techniques. Moreover, when hyperspectral measurements are taken at different temporal frequencies, separation is likely to improve; however, additional complexities in modelling time variability concurrently with this high spectral dimensionality may be created. As a result, the applicability of ensemble-based techniques suitable for high-dimensional data is examined in this research, together with the statistical evaluation of time-induced variability, since spectral measurements of tree species were taken at different time periods. Classification errors for the stochastic gradient boosting (SGB) and random forest (RF) methods ranged between 5.6% and 13.5%, respectively. Differences in classification accuracy or errors were also accounted for in the assessment of the models, with up to 46% of variation in classification error due to the effect of time in the RF model, indicating that measurement time is important in improving discrimination between tree species. This is because optical leaf characteristics can vary during the course of the year due to seasonal effects, health status, or the developmental stage of a tree. Different spectral properties (assumed from relevant wavelength bands) were found to be key factors impacting the models’ discrimination performance at various measurement times
Assessment of accuracy: systematic reduction of training points for maximum likelihood classification and mixture discriminant analysis (Gaussian and t-distribution)
No full textRemote sensing provides a valuable tool for monitoring land cover across large areas of land. A simple yet popular method for land cover classification is Maximum Likelihood Classification (MLC), which assumes a single normal distribution of the samples per class in the feature space. Mixture Discriminant Analysis (MDA) is a natural extension of MLC which can be used with varying distributions and multiple distributions per class, which simplifies the classification process tremendously. We compare the accuracies of MLC and MDA (using a Gaussian and t-distribution) as the number of training points are systematically reduced in order to simulate varying reference data availability conditions. The results show that the more robust t-distribution MDA performs comparatively with the Gaussian MDA and that both outperform MLC when sufficient training points are available. As the number of training points increases the MDA accuracies increase while the MLC accuracy stagnates. At very low numbers of training samples (ranging from 22 to 169 dependent on the class), there is more variability in terms of which method performs best
Ensemble classification for identifying neighbourhood sources of fugitive dust and associations with observed PM 10
No full textIn urban areas the deterioration of air quality as a result of fugitive dust receives less attention than the more prominent traffic and industrial emissions. We assessed whether fugitive dust emission sources in the neighbourhood of an air quality monitor are predictors of ambient PM10 concentrations on days characterized by strong local winds. An ensemble maximum likelihood method is developed for land cover mapping in the vicinity of an air quality station using SPOT 6 multi-spectral images. The ensemble maximum likelihood classifier is developed through multiple training iterations for improved accuracy of the bare soil class. Five primary land cover classes are considered, namely built-up areas, vegetation, bare soil, water and ‘mixed bare soil’ which denotes areas where soil is mixed with either vegetation or synthetic materials. Preliminary validation of the ensemble classifier for the bare soil class results in an accuracy range of 65–98%. Final validation of all classes results in an overall accuracy of 78%. Next, cluster analysis and a varying intercepts regression model are used to assess the statistical association between land cover, a fugitive dust emissions proxy and observed PM10. We found that land cover patterns in the neighbourhood of an air quality station are significant predictors of observed average PM10 concentrations on days when wind speeds are conducive for dust emissions. This study concludes that in the absence of an emissions inventory for ambient particulate matter, PM10 emitted from dust reservoirs can be statistically accounted for by land cover characteristics. This supports the use of land cover data for improved prediction of PM10 at locations without air quality monitoring stations
- …
