A Method for Palaeoclimatic Reconstruction in Palynology Based on Multivariate Time-Series Analysis

Much progress has been made in recent years in multivariate time-series analysis. In this paper we summarize some of the methodological developments that are particularly relevant to empirical economics and highlight especially the usefulness of linear transformations in analyzing multivariate time series. The topics considered include vector ARMA models, principal component analysis, scalar component models, canonical correlation analyses, co-integration, and unit-root tests. We illustrate the methods considered by an example using Taiwan’s interest-rate series and provide critiques of these developments.

Towards efficient state of charge estimation of lithium-ion batteries using canonical correlation analysis

Combining wind farms with concentrating solar plants to provide stable renewable power

Recent pollen spectra and zonal vegetation in the western USSR

Time Series: State Space Methods

Using a hypothetical data set, the authors provide concrete examples to illustrate that canonical correlation analysis is a general linear model, subsuming other parametric procedures as special cases. Specific statistical techniques included in the analysis are t tests, Pearson correlation, multiple regression, ANOVA, MANOVA, and discriminant analysis. The discussion is aided by an initial explanation of the logic of canonical analysis. Further, similarities between the canonical technique and other univariate and multivariate procedures are highlighted. The treatment is intended to reiterate a framework in which statistical concepts can be presented to students.

The Catholic Education Office (CEO) Sydney is a large non-government education authority which administers the systemic, Catholic schools of the Archdiocese of Sydney, Australia. The system consists of 148 primary and secondary schools with an enrolment of some 62,000 students. The major research question was: What characteristics of a learning organization can be identified in the Catholic Education Office (CEO) Sydney and are these perceived to raise standards in systemic schools of the Archdiocese of Sydney? Like all western education systems the CEO Sydney is immersed in constant change and is expected to account for improving educational standards within the system. The learning organization with its emphasis on adaptability and continuous improvement was considered an appropriate framework within which to conduct this research. The study consisted of two main parts the first investigated the CEO Sydney as a learning organization using a survey questionnaire distributed, using a dedicated web site, to a sample of primary and secondary principals in the system and a smaller number of senior CEO Sydney personnel. The response rate was 91%. This was complemented by examination of relevant CEO Sydney documentation and policies. The definition of the learning organization adopted for the study consisted of eight characteristics each of which formed a scale in the questionnaire. The eight characteristics adopted were: 'Systemic Thinking and Mental Models', 'Continuous Improvement of Work', 'Taking Initiatives and Risks', 'Ongoing Professional Development', 'Trusting and Collaborative Climate', 'Shared and Monitored Vision/Mission', 'Effective Communication Channels' and 'Team Work and Team Learning'. This part of the study was essentially a quantitative one, with the data subjected to descriptive, statistical analysis complemented by some clarifying and contextualising qualitative data. The second part of the study investigated the perceived relationship between the CEO Sydney and its learning organization characteristics and the standards in three curriculum outcome areas (religious education, literacy and numeracy). This part of the study was also quantitative using descriptive statistics complemented by Pearson correlation, multiple regression and canonical correlational analyses. Once again some relevant contextualising qualitative data was gathered. Five demographic groups (gender, role, region (principals only), years of experience as a principal and age) were examined to see if there were any differences in the extent to which the various learning organization characteristics and curriculum outcomes were identified by each group. The results of this study indicated that the CEO Sydney exhibited many of the characteristics of a learning organization with particular strengths in 'Continuous Improvement of Work', 'Systemic Thinking and Mental Models' and 'Shared and Monitored Vision/Mission'. The weakest characteristic was 'Taking Initiatives and Risks'. Demographic group analysis of this data revealed that there were no statistically significant differences in the responses of the different demographic groups. The results also indicated that there were correlations between the CEO Sydney as a learning organization and raising standards particularly in religious education and literacy and less so in numeracy. Finally, the study made a number of recommendations for the further development of the CEO Sydney as a learning organization and ways that it can further raise standards in the schools of the system.

SUMMARY We propose a method to specify an appropriate yet parsimonious vector autoregressive moving average (ARMA) model for a given multivariate time series. By considering contemporaneous linear transformations of the vector process, we introduce the concept of scalar component models within the vector ARMA framework (a) to reveal possibly hidden simplifying structures of the process, (b) to achieve parsimony in parameterization and (c) to identify the exchangeable models. The simplifying structures are of particular importance in the analysis of multivariate time series because they are often not obvious from the observed data but can be used to gain insights to the process under study. The analytical tool used to search for scalar component models is a canonical correlation analysis of vector processes and the proposed procedures are illustrated via two real examples.

The purpose of this study is to explore the correlations among knowledge Management strategy, school-based management and school effectiveness. It includes; (a) to investigate the present status of knowledge management strategy, school-based management effectiveness; (b) to compare the differences among the knowledge management strategy, school-based management, school effectiveness; (c) to probe into the correlation between knowledge management strategy and school-based management, knowledge management strategy and school effectiveness, school-based management and school effectiveness. Some of 44 public elementary schools in Tainan are selected as samples. One way ANOVA, Canonical Correlation, Multiple Regression, Cluster analysis, Discriminant analysis and T-test are applied on the data obtained from questionnaires of the 44 schools. The results show that those schools adopted codificational knowledge management strategy have positive influence on both school-based management ability and school effectiveness. Those schools adopted codificational knowledge management strategy are shown to be better on both school-based management ability and school effectiveness.

With the rapid development of the WoT, acquisition requirements to the data grow with each passing day. However, the existing data acquisition systems usually use the upload mode. Uploading data not only takes a lot of time, but also requires staff to continuously monitor on duty. Aiming at the problem of enterprise management data acquisition, a WoT-based enterprise management data acquisition and analysis system is constructed. The system obtains multiple monitoring data through the data acquisition module and integrates these data through the processing module, which can solve the problems of traditional data acquisition systems. In order to effectively fuse data, based on canonical correlation analysis (CCA), the multiple order statistics correlation discrimination analysis algorithm is proposed as a fusion strategy. Experimental results show that the fusion strategy based on CCA can effectively integrate monitoring data and accurately identify different characters.

Poyang Lake is the largest freshwater lake in China. Because the lake is faced with a series of ecological and environmental problems, it is important and necessary to study the land use changes in the Poyang Lake region. In this study, Landsat images from 1976, 1989, 1999 and 2009, were used along with other evaluation methods (quantitative analysis of changes in land use, land use dynamic, degree and transfer matrix analyses) to investigate the changes in land use within the Poyang Lake region from 1976 to 2009. Land use was classified into seven types: cropland, woodland, grassland, the water body, construction land, bottomland and unused land. Areas of cropland, bottomland and the water body were decreasing although the area of the water body decreased relatively slowly. However, areas of construction land, grassland, woodland and unused land increased to a certain extent, with unused land increasing the most rapidly given its smaller proportion compared to the other types. All land types in the study region had different levels of conversion between other types of land. As a whole, the decrease in cropland and the increase in construction land, woodland and grassland were caused by human activities and the conservation policies such as Grain-for-Green and “returning farm to lake”. However, the decrease in the water body area was mainly influenced by natural factors and human activities. Changes in the area of bottomland were mainly influenced by water levels and human farming activities.

As human activities in natural areas increase, understanding human–wildlife interactions is crucial. Big data approaches, like large‐scale camera trap studies, are becoming more relevant for studying these interactions. In addition, open‐source object detection models are rapidly improving and have great potential to enhance the image processing of camera trap data from human and wildlife activities. In this study, we evaluate the performance of the open‐source object detection model MegaDetector in cross‐regional monitoring using camera traps. The performance at detecting and counting humans, animals and vehicles is evaluated by comparing the detection results with manual classifications of more than 300 000 camera trap images from three study regions. Moreover, we investigate structural patterns of misclassification and evaluate the results of the detection model for typical temporal analyses conducted in ecological research. Overall, the accuracy of the detection model was very high with 96.0% accuracy for animals, 93.8% for persons and 99.3% for vehicles. Results reveal systematic patterns in misclassifications that can be automatically identified and removed. In addition, we show that the detection model can be readily used to count people and animals on images with underestimating persons by −0.05, vehicles by −0.01 and animals by −0.01 counts per image. Most importantly, the temporal pattern in a long‐term time series of manually classified human and wildlife activities was highly correlated with classification results of the detection model (Pearson's r = 0.996, p < 0.001) and diurnal kernel densities of activities were almost equivalent for manual and automated classification. The results thus prove the overall applicability of the detection model in the image classification process of cross‐regional camera trap studies without further manual intervention. Besides the great acceleration in processing speed, the model is also suitable for long‐term monitoring and allows reproducibility in scientific studies while complying with privacy regulations.

Warming trends are responsible for an observed decrease of water discharge in Southern France (northwestern Mediterranean). Ongoing climate change and the likely increase of water demand threaten the availability of water resources over the coming decades. Drought indices like the Reconnaissance Drought Index (RDI) are increasingly used in climate characterization studies, but little is known about the relationships between these indices, water resources and the overall atmospheric circulation patterns. In this study, we investigate the relationships between the RDI drought index, water discharge and four atmospheric teleconnection patterns (TPs) for six coastal river basins in southern France, both for the historical period of the last 60 years and for a worst-case climatic scenario (RCP 8.5) reaching the year 2100. We combine Global and Regional Climate Model (CGM and RCM, respectively) outputs with a set of observed climatic and hydrological data in order to investigate the past relationships between RDI, water discharge and TPs and to project their potential evolutions in space and time. Results indicate that annual water discharge can be reduced by −49/−88 % by the end of the century under the extreme climate scenario conditions. Due to unequal links with TPs, the hydro-climatic evolution is unevenly distributed within the study area. Indeed a clustering analysis performed with the RDI time series detects two major climate clusters, separating the eastern and western part of the study region. The former indicates stronger relationships with the Atlantic TPs (e.g. the NAO and the Scand patterns) whereas the latter is more closely related to the Mediterranean TPs (MO and WeMO). The future climate simulations predict an antagonistic evolution in both clusters which are likely driven by decreasing trends of Scand and WeMO. The former provokes a general tendency of lower P in both clusters during spring, summer and autumn, whereas the latter might partly compensate this evolution in the eastern cluster during autumn and winter. However, compared to observations, representation of the Mediterranean TPs WeMO and MO in the considered climate models is less satisfactory compared to the Atlantic TPs NAO and Scand, and further improvement of the model simulations therefore requires better representations of the Mediterranean TPs.

Soil salinity is a widespread natural hazard that negatively influences soil fertility and crop productivity. Using the potential of earth observation data and remote sensing technologies provides an opportunity to address this environmental issue and makes it possible to identify salt-affected regions accurately. While most of the utilized methods and model development techniques for monitoring soil salinity to date have been globally considered and tried to detect salinity and create predictive maps with a single regression algorithm, fewer studies have investigated the potential of local models and weighted regression techniques for estimating soil salinity. Accordingly, this research deals with monitoring surface soil salinity by the potential of Sentinel-2 multispectral imagery using the geographic weighted regression (GWR) technique. The field study was conducted in an area that has suffered from salinization, and the salinity of several soil samples was measured to be used as a source of ground truth data. The most efficient satellite features, which accurately predict surface soil salinity by its higher spectral reflectance, were derived from the Sentinel-2 data to be used as explanatory variables in the analysis. The GWR algorithm was then implemented with a fixed Gaussian kernel, and the optimized bandwidth was calculated in a calibration process using the cross-validation score (CV score). The results of the analysis proved that the GWR method has a great capability to predict soil salinity with an accuracy of two decimal places. The visual interpretation of the local estimates of coefficients and local t-values for each predictor variable has also been provided, which highlights the local variations in the study site. Finally, the achieved results were compared with the outcomes obtained from implementing two global regression techniques, Support Vector Machines (SVM), and Multiple Linear Regression (MLR), which confirmed the higher performance of the GWR algorithm.

Breast cancer is one of the major life-threatening diseases in the world today. But early diagnosis would ensure timely treatment and even recovery. Apart from the medical procedures that help to identify cancer, techniques such as machine learning can be applied to predict the occurrence of the disease in patients by analyzing their cell data. In this work, an attempt is made to apply three different regression techniques Multiple, Ridge, and Polynomial regression techniques. The performance metrics reveal that Multiple regression has a better performance compared to the other models.