How we raise young children is one of today's most highly personalized and sharply politicized issues, in part because each of us can claim some level of "expertise." The debate has intensified as discoveries about our development-in the womb and in the first months and years-have reached the popular media. How can we use our burgeoning knowledge to assure the well-being of all young children, for their own sake as well as for the sake of our nation? Drawing from new findings, this book presents important conclusions about nature-versus-nurture, the impact of being born into a working family, the effect of politics on programs for children, the costs and benefits of intervention, and other issues. The committee issues a series of challenges to decision makers regarding the quality of child care, issues of racial and ethnic diversity, the integration of children's cognitive and emotional development, and more. Authoritative yet accessible, From Neurons to Neighborhoods presents the evidence about "brain wiring" and how kids learn to speak, think, and regulate their behavior. It examines the effect of the climate-family, child care, community-within which the child grows.

This volume in the Challenges in Machine Learning series gathers papers from the Mini Symposium on Causality in Time Series, which was part of the Neural Information Processing Systems (NIPS) confernce in 2009 in Vancouver, Canada. These papers present state-of-the-art research in time-series causality to the machine learning community, unifying methodological interests in the various communities that require such inference.

A state of the art volume on statistical causality Causality: Statistical Perspectives and Applications presents a wide-ranging collection of seminal contributions by renowned experts in the field, providing a thorough treatment of all aspects of statistical causality. It covers the various formalisms in current use, methods for applying them to specific problems, and the special requirements of a range of examples from medicine, biology and economics to political science. This book: Provides a clear account and comparison of formal languages, concepts and models for statistical causality. Addresses examples from medicine, biology, economics and political science to aid the reader's understanding. Is authored by leading experts in their field. Is written in an accessible style. Postgraduates, professional statisticians and researchers in academia and industry will benefit from this book.

Interrupted Time Series Analysis develops a comprehensive set of models and methods for drawing causal inferences from time series. It provides example analyses of social, behavioral, and biomedical time series to illustrate a general strategy for building AutoRegressive Integrated Moving Average (ARIMA) impact models. Additionally, the book supplements the classic Box-Jenkins-Tiao model-building strategy with recent auxiliary tests for transformation, differencing, and model selection. Not only does the text discuss new developments, including the prospects for widespread adoption of Bayesian hypothesis testing and synthetic control group designs, but it makes optimal use of graphical illustrations in its examples. With forty completed example analyses that demonstrate the implications of model properties, Interrupted Time Series Analysis will be a key inter-disciplinary text in classrooms, workshops, and short-courses for researchers familiar with time series data or cross-sectional regression analysis but limited background in the structure of time series processes and experiments.

This monograph presents a brief overview of the literature on the difference-in-difference estimation strategy and discusses major issues mainly using a treatment effect perspective that allows more general considerations than the classical regression formulation that still dominates the applied work.

Explains the value of quasi-experimental techniques that can be used to approximate randomized experiments. The goal is to describe the logic of causal inference for researchers and policymakers who are not necessarily trained in experimental and quasi-experimental designs and statistical techniques.

Extensive code examples in R, Stata, and Python Chapters on overlooked topics in econometrics classes: heterogeneous treatment effects, simulation and power analysis, new cutting-edge methods, and uncomfortable ignored assumptions An easy-to-read conversational tone Up-to-date coverage of methods with fast-moving literatures like difference-in-differences

The central aim of many studies in population research and demography is to explain cause-effect relationships among variables or events. For decades, population scientists have concentrated their efforts on estimating the ‘causes of effects’ by applying standard cross-sectional and dynamic regression techniques, with regression coefficients routinely being understood as estimates of causal effects. The standard approach to infer the ‘effects of causes’ in natural sciences and in psychology is to conduct randomized experiments. In population studies, experimental designs are generally infeasible. In population studies, most research is based on non-experimental designs (observational or survey designs) and rarely on quasi experiments or natural experiments. Using non-experimental designs to infer causal relationships—i.e. relationships that can ultimately inform policies or interventions—is a complex undertaking. Specifically, treatment effects can be inferred from non-experimental data with a counterfactual approach. In this counterfactual perspective, causal effects are defined as the difference between the potential outcome irrespective of whether or not an individual had received a certain treatment (or experienced a certain cause). The counterfactual approach to estimate effects of causes from quasi-experimental data or from observational studies was first proposed by Rubin in 1974 and further developed by James Heckman and others. This book presents both theoretical contributions and empirical applications of the counterfactual approach to causal inference.

Time-varying data are prevalent in a wide variety of real-world applications for example health care, environmental study, finance, motion capture among others. Time-varying data possess complex nature and pose unique challenges. For example, time-varying data observed in real-world applications almost always exhibit nonstationary characteristics that challenges ordinary time-series methods with stationary assumptions. In addition, one may only have access to irregularly sampled data which prohibits the models that assume regularly observed samples. On the other hand, as machine learning and data mining algorithms have begun make an impact on real-world applications, merely providing accurate prediction is no longer sufficient. There is a growing need for interpretations and explanations to how the machine learning models make predictions in order for end-users to fully trust and adopt these models. In this thesis, we explore time-varying data in various practical scenarios and aim at enhancing model explainability and understanding of the data. First, we study the problem of building explainable classifiers for multivariate time series data by means of joint variable and time interval selection. We introduce a modular framework, the LAXCAT model, consisting of a convolution-based feature extraction and a dual attention mechanism. The convolution-based feature extraction network produces variable-specific representation by considering local time interval context. The dual attention mechanisms, namely variable attention network and temporal attention network, work in concert to simultaneously select variable and time interval that are discriminative to the classification task. We present results of extensive experiments with several benchmark data sets that show that the proposed method outperforms the state-of-the-art baseline methods on multi-variate time series classification task. The results of our case studies demonstrate that the variables and time intervals identified by the proposed method make sense relative to available domain knowledge. Second, to obtain a better understanding of the input multivariate time series data, we study dynamic structure learning which aims at jointly discovering hidden state transitions and state-dependent inter-variable connectivity structures. To address the research problem, we introduce a novel state-regularized dynamic autoregressive model framework, the SrVARM model, featuring a state-regularized recurrent neural network and a dynamic autoregressive model. The state-regularized recurrent unit learns to discover the hidden state transition dynamics from the data while the autoregressive function learns to encode state-dependent inter-variable dependencies in directed acyclic graph structure. A smooth characterization of the acyclic constraint is exploited to train the model in an efficient and unified framework. We report results of extensive experiments with simulated data as well as a real-world benchmark that show that SrVARM outperforms state-of-the-art baselines in recovering the unobserved state transitions and discovering the state-dependent relationships among variables. Third, functional data analysis provides another promising perspective at dealing with time-vary data. However, the representation learning capability of neural network-based method have not been fully explored for functional data. We study unsupervised representation learning from functional data and introduce the functional autoencoder network which generalizes the standard autoencoder network to the functional data setting. The functional autoencoder copes with functional data input by leveraging functional weights and inner product for real-valued functions. We derive from first principles, a functional gradient-based algorithm for training the resulting network. We present results of experiments which demonstrate that the functional autoencoders outperform the state-of-the-art baseline methods. Besides providing a solution to the problem of functional data representation learning, the proposed model offers a fundamental building block for other functional data learning tasks, such as classification and regression networks. Fourth, we study the problem of treatment effect estimation from networked time series data. Such data arise in settings where individuals are linked by a network of relations, e.g., social ties, and the observations for each individual are naturally represented by time series. We propose a novel representation learning approach to treatment effect estimation from networked time series data consisting of a temporal convolution network, a graph attention network, and a treatment-specific outcome predictor network. We use an adversarial learning framework for domain adaptation to learn a representation of individuals that makes treatment assignment independent of the treatment outcome. We present results of experiments and show that the proposed framework outperforms the state-of-the-art baselines in estimating treatment effects from networked time series data. We conclude with a brief summary of the main contributions of the thesis and some directions for further research.