Are Statisticians Sufficiently Engaged With Public Policy?

Agriculture is an important tool in poverty alleviation especially in rural areas whose economy mainly relies on agriculture as a source of income. Nevertheless, Nigeria is to face the food security crisis due to its agriculture sector which it heavily depends, but food consumption is imported rather than self-produced. Projections indicate that by 2030 Nigeria’s population will double the 2006 estimation which necessitates increased food production to meet the growing and urbanizing population needs and to open possibilities of exports. Apart from this, the country is faced with soil degradation as a result of inappropriate agricultural practices; erosion and gully, deforestation as well as climate change. It endangers the once dominant subsistent farm economy. The problems rose above show that there is an urgent need to review the role of fertilizers in food production for better food security. Proper execution of fertilizer policies becomes a powerful tool towards increasing agricultural production, malnutrition reduction and poverty alleviation through lowering food prices. Nevertheless, demand and supply factors like low farmers’ incomes, high market prices due the limited availability of fertilizer and public policy response also known as price incentives influence the low usage of fertilizers in Nigeria. In view of above problems, there is an urgent need to have a more holistic approach to sustainably raise agricultural productivity in Nigeria with a view to averting food shortage. Keywords: Food security, Fertilizer production, Agricultural productivity, Poverty alleviation, Policy frameworks

Some Ghosts of Statistics Past

Ground deformation caused by groundwater exploitation leads to significant socio-economic losses worldwide. Driving factors such as population growth and climate change will increase these losses, especially in arid regions where droughts are becoming more intense, longer lasting, and frequent. Therefore, there is a need to generate models capable of forecasting ground deformation. However, few studies have analyzed deformation time series (DTS) to identify and characterize subsidence phenomena.Our research aims to predict the ground deformation associated with groundwater abstractions in 18 wells of the Madrid Detrital Aquifer (ATDM) using statistical models and shallow and deep Machine Learning (ML) algorithms. We generated a database with 18 monthly time series (one for each well) between 1992 and 2010, with data for two variables: a binary variable indicating extraction-recovery cycles of the aquifer and a continuous variable representing the average deformation for the area of influence of each well. DTS generated from Persistent Scatter Interferometry (PSI) of ERS-1/2 and ENVISAT radar images were used to calculate the average deformation. Finally, we applied six different methods for forecasting DTS: two statistical models, Autoregressive Integrated Moving Average (ARIMA) and Prophet (P), one ensemble shallow ML algorithm, Random Forest (RF), one hybrid method, Neural Prophet (NP), and two Deep Learning (DL) techniques 1D Convolutional Neural Networks (CNN1D), and Long Short-Term Memory (LSTM).The analysis of DTS allowed us to differentiate two zones with different hydrological behavior: a zone of higher permeability (north zone) and another of lower permeability (south zone). We found that establishing the architectures of ML and DL algorithms based on hydrological zones improves the prediction of ground deformation. ML and DL algorithms provide better forecasts compared to statistical and hybrid models. Specifically, LSTM and RF offer the best results. Our results show the potential of LSTM algorithms and the previous grouping of DTS in predicting ground deformation associated with groundwater exploitation.This work has been developed thanks to the pre-doctoral grant for the Training of Research Personnel (PRE2021-100044) funded by MCIN/AEI/10.13039/501100011033 and by "FSE invests in your future" within the framework of the SARAI project "Towards a smart exploitation of land displacement data for the prevention and mitigation of geological-geotechnical risks" PID2020-116540RB-C22 funded by MCIN/AEI/10.13039/501100011033.

Chapter 6 - Machine learning for soil moisture assessment

Can we live within environmental limits and still reduce poverty? Degrowth or decoupling?

Although there is a relationship between climate change and poverty reduction, some countries in the Sahel region of Africa are yet to incorporate a climate change risk management strategy into their poverty reduction programme. The aim of this research is therefore to evaluate the subjective and quantitative analysis of climate change risk management strategies in all the nine countries that constitute the Sahel region. Both a qualitative and a quantitative approach were used in the investigation. The qualitative approach was used to make a subjective analysis of the climate change mitigating framework with regard to poverty reduction issues at country level. It was based on a scoring system methodology by assessing the following criteria: The consideration of climate change scenarios and the vulnerabilities of the countries; The analysis of poverty-climate links; The climate change institutional framework of each country. The quantitative approach was based on Data Envelopment Analysis (DEA). This was done by applying DEA Solver to evaluate the efficiency of climate change mitigating frameworks and their impact on poverty reduction. The results of the investigation revealed that none of the Sahel countries, excluding Burkina Faso, included climate change risk management in their Poverty Reduction Strategy and Policies (PRSP). Burkina Faso’s National Adaptation Policies Action (NAPA) performs well. Burkina Faso is considered a model country which has included climate change policies in its PRSP. However, as with most countries in the region, the Burkina Faso climate change risk management policy is not comprehensive. This is a result of inadequate climate change risk management projects, exclusion of real needs and other exogenous parameters. However, in Senegal, more efforts are required in response to climate change risk management in order to reduce poverty. This is paramount for the following reasons: the inclusion of climate change risk management into PRSP important in increasing its understanding several challenges require to be addressed for good implementation and to ensure a good monitoring system. Stakeholders in the sector will be more effective and efficient in developing action plans for climate change management strategies and poverty reduction programmes.KeywordsSahelClimate changePoverty reductionAdaptation strategyBiosketches

Plants meet machines: Prospects in machine learning for plant biology

This dissertation consists of three independent, self-contained research papers that investigate how state-of-the-art machine learning algorithms can be used in combination with operations management models to consider high dimensional data for improved planning decisions. More specifically, the thesis focuses on the question concerning how the underlying decision support models change structurally and how those changes affect the resulting decision quality. Over the past years, the volume of globally stored data has experienced tremendous growth. Rising market penetration of sensor-equipped production machinery, advanced ways to track user behavior, and the ongoing use of social media lead to large amounts of data on production processes, user behavior, and interactions, as well as condition information about technical gear, all of which can provide valuable information to companies in planning their operations. In the past, two generic concepts have emerged to accomplish this. The first concept, separated estimation and optimization (SEO), uses data to forecast the central inputs (i.e., the demand) of a decision support model. The forecast and a distribution of forecast errors are then used in a subsequent stochastic optimization model to determine optimal decisions. In contrast to this sequential approach, the second generic concept, joint estimation-optimization (JEO), combines the forecasting and optimization step into a single optimization problem. Following this approach, powerful machine learning techniques are employed to approximate highly complex functional relationships and hence relate feature data directly to optimal decisions. The first article, “Machine learning for inventory management: Analyzing two concepts to get from data to decisions”, chapter 2, examines performance differences between implementations of these concepts in a single-period Newsvendor setting. The paper first proposes a novel JEO implementation based on the random forest algorithm to learn optimal decision rules directly from a data set that contains historical sales and auxiliary data. Going forward, we analyze structural properties that lead to these performance differences. Our results show that the JEO implementation achieves significant cost improvements over the SEO approach. These differences are strongly driven by the decision problem’s cost structure and the amount and structure of the remaining forecast uncertainty. The second article, “Prescriptive call center staffing”, chapter 3, applies the logic of integrating data analysis and optimization to a more complex problem class, an employee staffing problem in a call center. We introduce a novel approach to applying the JEO concept that augments historical call volume data with features like the day of the week, the beginning of the month, and national holiday periods. We employ a regression tree to learn the ex-post optimal staffing levels based on similarity structures in the data and then generalize these insights to determine future staffing levels. This approach, relying on only few modeling assumptions, significantly outperforms a state-of-the-art benchmark that uses considerably more model structure and assumptions. The third article, “Data-driven sales force scheduling”, chapter 4, is motivated by the problem of how a company should allocate limited sales resources. We propose a novel approach based on the SEO concept that involves a machine learning model to predict the probability of winning a specific project. We develop a methodology that uses this prediction model to estimate the “uplift”, that is, the incremental value of an additional visit to a particular customer location. To account for the remaining uncertainty at the subsequent optimization stage, we adapt the decision support model in such a way that it can control for the level of trust in the predicted uplifts. This novel policy dominates both a benchmark that relies completely on the uplift information and a robust benchmark that optimizes the sum of potential profits while neglecting any uplift information. The results of this thesis show that decision support models in operations management can be transformed fundamentally by considering additional data and benefit through better decision quality respectively lower mismatch costs. The way how machine learning algorithms can be integrated into these decision support models depends on the complexity and the context of the underlying decision problem. In summary, this dissertation provides an analysis based on three different, specific application scenarios that serve as a foundation for further analyses of employing machine learning for decision support in operations management.

Climate change has become one of the most critical problems threatening our world, gaining increased attention in either academia or industry. Climate change has been demonstrated as the major barrier in the way of sustainable development strategy in the 2030 Agenda. Nowadays, the Social Internet of Things (SIoT) has paved new ways for public deliberations and has transformed the communication of global issues such as climate change. Thus, sentiment analysis of SIoT media streams can offer great help in improving the mitigation and adaptation to climate change. Machine learning (ML) is demonstrating great success in a wide range of SIoT applications. However, training ML algorithms for sentimental analysis of climate change is notoriously hard as it suffers from feature engineering issues, information squashing, unbalancing, and curse-of-dimensionality, which bounds their possible power for modeling social awareness of climate change. Besides, the absence of a standard benchmark with reasonable and dependable experimentations brings a practically intractable difficulty to the evaluation of the efficiency of new solutions. In this regard, this study introduces the first reasonable and reproducible benchmark devoted to evaluating the potential of ML algorithms in identifying users’ opinions about climate change. Moreover, a novel taxonomy is presented for categorizing the existing ML algorithms, exploring their optimal hyperparameter, and unifying their elementary settings. Inclusive experiments are then performed on real Twitter data with different families of ML algorithms. To promote further study, a detailed analysis is provided for the state of the field to uncover the open research challenges and promising future directions.

Climate change through extreme weather events threatens food security (FS) and the eradication of poverty. Thus, improving FS will require adapting to the impacts of climate change as well as reducing the risks of disasters. However, the nexus between FS, disaster risk reduction (DRR) and climate change adaptation (CCA) is not always reflected in policies, resulting in fragmented implementation. The purpose of this article is to evaluate if there is coherence in the policies for FS, DRR and CCA in South Africa. A qualitative research design was applied, and data were collected through a summative content analysis on 34 policy and legislative documents and 24 key informant interviews (KII). The study found that there are still incoherencies between the current main policy and legislative documents that address CCA, DRR and FS. This study recommends a review of old policy and legislative frameworks promulgated in the 1990s to incorporate cross-cutting issues such as DRR, CCA and FS. This will enhance and strengthen synergies and interconnections between the three policy areas.

“Good luck on your first day as an assistant professor, Dr. Tanner! Have a great class!” On the wall above my desk, these words scream out from an otherwise encouraging note that is adorned with many exclamation points. This note has hung on my wall since my very first day as an Assistant Professor of Biology. As I was charging off to teach my first class, a senior faculty member who had been on my hiring committee slipped this note under my office door. In moments of pause years later, I still stare up at that note and breathe a sigh of relief that I had much more than luck to guide me on my first day as a college-level teacher. Although I continue to have much to learn—as all of us do no matter the number of years of teaching experience—I did arrive at the university with both formal and informal training in science education. I had had plenty of exposure to innovative pedagogical approaches, questioning strategies, and techniques for engaging diverse audiences in learning science. As a scientist educator, I had had the privilege of many years of collaboration with outstanding K–12 educators as well as a postdoctoral fellowship in science education. However, my training has been, to say the least, unconventional compared with that of my fellow junior faculty and unique in its preparation in regard to the teaching and learning of my discipline. It will not be news to anyone reading this article that university and college teaching is to a large extent a profession with no formal training. It’s startling but true that the majority of faculty members—and lecturers who often teach large numbers of students—have no formal training in the teaching and learning of their discipline. In fact, the hiring process in university science departments is structured primarily to evaluate a faculty candidate’s ability to be a productive researcher, with success measured in number of publications and magnitude of grant funds raised. Depending on the type of institution, for example, research university, state-level university, or liberal arts college, there may be a component of the faculty interview process that probes a candidate’s teaching ability, for example, requesting a statement of teaching philosophy and requiring the candidate to teach a sample lecture class. However, this sample lecture often screens for gross inadequacies, rather than looking for stellar innovations or pedagogical skills. This lack of formal, accredited training for university and college instructors stands in stark contrast to the requirements for a high school teacher who is charged with the education of students only a year junior to college freshmen. High school teachers in the United States must be credentialed as a secondary science teacher, demonstrate subject matter competency in every subject that they will be teaching, and must continually engage in professional development in the teaching and learning of their discipline throughout their career as a science teacher. With the 2002 federal No Child Left Behind legislation, the onus is upon each precollege science teacher to become “highly qualified” in terms of formal university-level training in science education. However, no such required professional training or measurable standards for teaching are required in institutions of higher education. Many policy documents have suggested standards of teaching practice in postsecondary science education (National Research Council, 1996, 1997; Siebert and McIntosh, 2001), but the extent of implementation of these ideals is unclear and has gone relatively unstudied, although national and regional accreditation boards do look at outcomes, asking colleges and universities to assess what their students have gained from four years of study at their institutions. Nonetheless, there is a striking reversal of accountability that happens when one crosses the precollege teaching to college-level teaching boundary (Table 1). During the K–12 school years, society expects K–12 teachers to be responsible for student learning. Salaries of teachers in many states are tied to student test scores, and poor student performance can potentially invoke penalties. At a college or university, several variables in the educational universe shift. Students are the ones responsible for learning. The evaluation and compensation of college-level teachers is not DOI: 10.1187/cbe.05–12–0132 Address correspondence to: Kimberly Tanner (kdtanner@sfsu.edu). CBE—Life Sciences Education Vol. 5, 1–6, Spring 2006

Climate change is a global issue that must be considered and addressed immediately. Many articles have been published on climate change mitigation and adaptation. However, new methods are required to explore the complexities of climate change and provide more efficient and effective adaptation and mitigation policies. With the advancement of technology, machine learning (ML) and deep learning (DL) methods have gained considerable popularity in many fields, including climate change. This paper aims to explore the most popular ML and DL methods that have been applied for climate change mitigation and adaptation. Another aim is to determine the most common mitigation and adaptation measures/actions in general, and in urban areas in particular, that have been studied using ML and DL methods. For this purpose, word frequency analysis and topic modeling, specifically the Latent Dirichlet allocation (LDA) as a ML algorithm, are used in this study. The results indicate that the most popular ML technique in both climate change mitigation and adaptation is the Artificial Neural Network. Moreover, among different research areas related to climate change mitigation and adaptation, geoengineering, and land surface temperature are the ones that have used ML and DL algorithms the most.

Land use and land cover (LULC) classification plays a significant role in the analysis of climate change, evidence-based policies, and urban and regional planning. For example, updated and detailed information on land use in urban areas is highly needed to monitor and evaluate urban development plans. Machine learning (ML) algorithms, and particularly ensemble ML models support transferability and efficiency in mapping land uses. Generalization, model consistency, and efficiency are essential requirements for implementing such algorithms. The transfer-ensemble learning approach is increasingly used due to its efficiency. However, it is rarely investigated for mapping complex urban LULC in Global South cities, such as India. The main objective of this study is to assess the performance of machine and ensemble-transfer learning algorithms to map the LULC of two metropolitan cities of India using Landsat 5 TM, 2011, and DMSP-OLS nightlight, 2013. This study used classical ML algorithms, such as Support Vector Machine-Radial Basis Function (SVM-RBF), SVM-Linear, and Random Forest (RF). A total of 480 samples were collected to classify six LULC types. The samples were split into training and validation sets with a 65:35 ratio for the training, parameter tuning, and validation of the ML algorithms. The result shows that RF has the highest accuracy (94.43%) of individual models, as compared to SVM-RBF (85.07%) and SVM-Linear (91.99%). Overall, the ensemble model-4 produces the highest accuracy (94.84%) compared to other ensemble models for the Kolkata metropolitan area. In transfer learning, the pre-trained ensemble model-4 achieved the highest accuracy (80.75%) compared to other pre-trained ensemble models for Delhi. This study provides innovative guidelines for selecting a robust ML algorithm to map urban LULC at the metropolitan scale to support urban sustainability.

The past three decades have seen an unprecedented increase in world living standards and a fall in poverty across many fundamental dimensions. Increased confidence in what was possible together with greater acceptance of moral responsibilities led to the adoption of the Millennium Development Goals (MDGs) at the turn of the century. They provided a real basis for international cooperation and development. In the Sustainable Development Goals (SDGs), agreed in September 2015, there is now a common platform for the next phase of the fight against poverty. The SDGs make it clear that environmental protection will be a key feature of this next phase and increasingly intertwined with poverty reduction. Thirteen of the 17 SDGs are directly concerned with the natural environment, climate or sustainability. Environment, climate and sustainability were not prominent in the MDGs. With hindsight we can now see that this was a mistake. A key factor in all this is climate change. Climate change is not the only environmental problem we face, nor is it the only threat to global prosperity. But climate change is unique in its magnitude and the vast risks it poses. It is a potent threat-multiplier for other urgent concerns, such as habitat loss, disease and global security (IPCC 2014) and puts at risk the development achievements of the past decades (World Bank 2016). If unchecked, climate change could fundamentally redraw the map of the planet, and where and how humans and other species can live. Climate change is also unique in the scale of the response that is needed. Reducing climate risks requires cooperation from all countries, developed and developing, to reorient their economic systems away from fossil fuels and harmful land-use practices. This reorientation is urgent. Our activities in the next two decades will determine whether our successes in development will be sustained or advanced, or whether they will be undermined or reversed in a hostile environment. The nature of the climate problem has implications for economic analysis. Economics has much to offer, and indeed continues to provide important insights, but there has been a dangerous tendency to force climate change into narrow existing ways of thinking. This must change. We need to construct theories and models that reflect the structure and scale of the problem and the contexts in which it occurs. Climate change also has implications for development policy. In the Paris Agreement, negotiated at the end of 2015,there is now an international platform through which global climate action can be advanced and coordinated. The Paris Agreement sets out a process through which the rise in global mean temperatures may be curtailed to 'well below' 2oC above pre-industrial levels and perhaps as low as 1.5oC. These are the central long-term objectives of the agreement. Meeting the Paris objectives requires sustained action over many decades. It also requires the reorientation of investment. At least US$ 100 trillion will be invested over the next two or three decades into buildings and urban infrastructure, roads, railways, ports and into new energy systems. It is imperative that these investment decisions are taken with climate change in mind. If they are there will be substantial benefits for development and poverty reduction – living spaces where we can move, breathe and be productive, better protection for fragile ecosystems, as well as the fundamental reduction of the risks of climate change. Putting the SDGs and Paris together, the agreements of 2015 have given us, for the first time, a global agenda for sustainable development applying to all countries. This paper sets out the implications of this agenda, and climate change in particular, for development economics and development policy. It emphasizes the nature of the required changes and their implications. We start with an examination of what economics has had to say about the link between economic prosperity and the environment. We then explain why climate change is a different kind of problem and why it requires a new approach to both analysis and policy. The final two sections explore how this new approach might look.

This chapter concentrates on approaches to environmental management and urban planning that have become skewed over time in order to resolve the African ‘environmental crisis’ in the name of sustainable development and/or poverty reduction. Such approaches have been brought to the fore by the global debate on Climate Change. The extent of the environmental transformations currently underway in Dar es Salaam and sub-Saharan Africa more generally are discussed, as well as the manner in which the two global strategies of mitigation and adaptation to Climate Change orient urban development policy and planning at the local and global level. The adaptation strategy emerges as crucial to planning processes in African cities, prompting a reconsideration of the impact of strategies that emphasize ‘securitization’ of the city as opposed to acceleration of the rural–urban transition in order to reduce social vulnerability. Such strategies raise questions that are not new to the planning debate, and draw attention to the role that people must play therein.