Time-series forecasting through modelling sequences of temporally dependent observations has many industrial and scientific applications. While machine learning models have been widely used to... Show moreTime-series forecasting through modelling sequences of temporally dependent observations has many industrial and scientific applications. While machine learning models have been widely used to create time-series forecasting models, creating efficient and performant time-series forecasting models is a complex task for domain users. Automated Machine Learning (AutoML) is a growing field that aims to make the process of creating machine-learning models accessible for non-machine learning experts. This is achieved by optimising machine learning pipelines automatically. Time-series machine-learning pipelines include various specialised pre-processing steps that are not currently supported by existing AutoML systems. This dissertation investigates how AutoML can be extended to time-series data analysis problems such as time-series forecasting. Several challenges arise when developing specialised AutoML systems for time-series forecasting. For instance, advanced machine-learning pipelines that can extract time-series features and select well-suited machine-learning models need to be developed. Also, extra hyperparameters such as the window size, which shows how many historical data points are helpful, need to be optimised by the AutoML system. This dissertation addresses these issues. We provide a comprehensive overview of the AutoML research field, including hyperparameter optimisation techniques, neural architecture search, and existing AutoML systems. Next, we investigate the use of AutoML for short-term forecasting, single-step ahead time-series forecasting, and multi-step time-series forecasting with time-series features. Show less
Over the last decades, income inequality has increased globally. How do social policies affect this increasing trend? How do international trade and technological progress affect inequality? What... Show moreOver the last decades, income inequality has increased globally. How do social policies affect this increasing trend? How do international trade and technological progress affect inequality? What is the profile of income inequality in China? Based on quantitative analyses of determinants of income inequality, this study provides a number of new insights into these questions. Income inequality has increased in the last decades all over the world. Several factors seem to contribute to this trend. Very prominent amongst them is the rising primary income inequality. The dominant income inequality-reducing effect comes from the tax benefit system, which offsets two thirds of the total increase in inequality. Generally speaking, the transition of welfare states from a traditional to a social investment oriented system does not lead to lower income inequality or poverty. There is also no robust and significant relationship between international trade and technology changes on the one hand, and income inequality on the other. Determinants of inequality in China are different from those in developed countries. In contrast to the tax benefit system in rich countries, the fiscal system in China does not bring a lower level of income inequality. Another explanation is the household registration system. Show less
This book focuses on novel approach to characterize developmental changes in pharmacokinetics across human lifespan with the up-to-date Nonlinear Mixed Effect Modeling techniques.