|
on Big Data |
| By: | Lachlan O'Neill (SoDa Laboratories, Monash Business School); Nandini Anantharama (SoDa Laboratories, Monash Business School); Wray Buntine (Faculty of Information Technology, Monash University); Simon D Angus (Dept. of Economics and SoDa Laboratories, Monash Business School) |
| Abstract: | Empirical social science requires structured data. Traditionally, these data have arisen from statistical agencies, surveys, or other controlled settings. But what of language, political speech, and discourse more generally? Can text be data? Until very recently, the journey from text to data has relied on human coding, severely limiting study scope. Here, we introduce natural language processing (NLP), a field of artificial intelligence (AI), and its application to discourse analysis at scale. We introduce AI/NLP’s key terminology, concepts, and techniques, and demonstrate its application to the social sciences. In so doing, we emphasise a major shift in AI/NLP technological capability now underway, due largely to the development of transformer models. Our aim is to provide the quantitative social scientists with both a guide to state-of-the-art AI/NLP in general, and something of a road-map for the transformer revolution now sweeping through the landscape. |
| Keywords: | text as data, artificial intelligence, machine learning, natural language processing, transformer models |
| JEL: | C45 C52 C55 |
| Date: | 2021–12 |
| URL: | http://d.repec.org/n?u=RePEc:ajr:sodwps:2021-12&r= |
| By: | Defina, Ryan |
| Abstract: | The field of deposit insurance is yet to realise fully the potential of machine learning, and the substantial benefits that it may present to its operational and policy-oriented activities. There are practical opportunities available (some specified in this paper) that can assist in improving deposit insurers’ relationship with the technology. Sharing of experiences and learnings via international engagement and collaboration is fundamental in developing global best practices in this space. |
| Keywords: | deposit insurance; machine learning |
| JEL: | G21 |
| Date: | 2021–09–15 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:110712&r= |
| By: | Luxuan Yang; Ting Gao; Yubin Lu; Jinqiao Duan; Tao Liu |
| Abstract: | With the fast development of modern deep learning techniques, the study of dynamic systems and neural networks is increasingly benefiting each other in a lot of different ways. Since uncertainties often arise in real world observations, SDEs (stochastic differential equations) come to play an important role. To be more specific, in this paper, we use a collection of SDEs equipped with neural networks to predict long-term trend of noisy time series which has big jump properties and high probability distribution shift. Our contributions are, first, we use the phase space reconstruction method to extract intrinsic dimension of the time series data so as to determine the input structure for our forecasting model. Second, we explore SDEs driven by $\alpha$-stable L\'evy motion to model the time series data and solve the problem through neural network approximation. Third, we construct the attention mechanism to achieve multi-time step prediction. Finally, we illustrate our method by applying it to stock marketing time series prediction and show the results outperform several baseline deep learning models. |
| Date: | 2021–11 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2111.13164&r= |
| By: | Feras A. Batarseh; Munisamy Gopinath; Anderson Monken; Zhengrong Gu |
| Abstract: | International economics has a long history of improving our understanding of factors causing trade, and the consequences of free flow of goods and services across countries. The recent shocks to the free trade regime, especially trade disputes among major economies, as well as black swan events, such as trade wars and pandemics, raise the need for improved predictions to inform policy decisions. AI methods are allowing economists to solve such prediction problems in new ways. In this manuscript, we present novel methods that predict and associate food and agricultural commodities traded internationally. Association Rules (AR) analysis has been deployed successfully for economic scenarios at the consumer or store level, such as for market basket analysis. In our work however, we present analysis of imports and exports associations and their effects on commodity trade flows. Moreover, Ensemble Machine Learning methods are developed to provide improved agricultural trade predictions, outlier events' implications, and quantitative pointers to policy makers. |
| Date: | 2021–11 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2111.07508&r= |
| By: | Faizal Hafiz; Jan Broekaert; Davide La Torre; Akshya Swain |
| Abstract: | This study proposes a new framework to evolve efficacious yet parsimonious neural architectures for the movement prediction of stock market indices using technical indicators as inputs. In the light of a sparse signal-to-noise ratio under the Efficient Market hypothesis, developing machine learning methods to predict the movement of a financial market using technical indicators has shown to be a challenging problem. To this end, the neural architecture search is posed as a multi-criteria optimization problem to balance the efficacy with the complexity of architectures. In addition, the implications of different dominant trading tendencies which may be present in the pre-COVID and within-COVID time periods are investigated. An $\epsilon-$ constraint framework is proposed as a remedy to extract any concordant information underlying the possibly conflicting pre-COVID data. Further, a new search paradigm, Two-Dimensional Swarms (2DS) is proposed for the multi-criteria neural architecture search, which explicitly integrates sparsity as an additional search dimension in particle swarms. A detailed comparative evaluation of the proposed approach is carried out by considering genetic algorithm and several combinations of empirical neural design rules with a filter-based feature selection method (mRMR) as baseline approaches. The results of this study convincingly demonstrate that the proposed approach can evolve parsimonious networks with better generalization capabilities. |
| Date: | 2021–11 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2111.08060&r= |
| By: | Annelore Verhagen |
| Abstract: | Technological developments are one of the major forces behind the need for retraining, but they can also be part of the solution. In particular, Artificial Intelligence (AI) has the potential to increase training participation, including among currently underrepresented groups, by lowering some of the barriers to training that people experience and by increasing motivation to train. Moreover, certain AI solutions for training may improve the alignment of training to labour market needs, and reduce bias and discrimination in the workplace. In order to realise the benefits of AI for training and ensure that it yields benefits for all, it will be necessary to address potential drawbacks in terms of changing skills requirements, inequalities in access to data, technology and infrastructure and important ethical issues. Finally, even when these drawbacks can be addressed, the introduction and expansion of AI tools for training is constrained by the supply of AI skills in the workforce and the availability of scientific evidence regarding the benefits of AI tools for training and whether they are cost-effective. |
| Keywords: | Adult learning, AI, Artificial Intelligence, Skills, Training |
| JEL: | I20 J24 M53 O15 O33 |
| Date: | 2021–12–13 |
| URL: | http://d.repec.org/n?u=RePEc:oec:elsaab:266-en&r= |
| By: | Kamesh Korangi; Christophe Mues; Cristi\'an Bravo |
| Abstract: | In this paper, we study mid-cap companies, i.e. publicly traded companies with less than US $10 billion in market capitalisation. Using a large dataset of US mid-cap companies observed over 30 years, we look to predict the default probability term structure over the medium term and understand which data sources (i.e. fundamental, market or pricing data) contribute most to the default risk. Whereas existing methods typically require that data from different time periods are first aggregated and turned into cross-sectional features, we frame the problem as a multi-label time-series classification problem. We adapt transformer models, a state-of-the-art deep learning model emanating from the natural language processing domain, to the credit risk modelling setting. We also interpret the predictions of these models using attention heat maps. To optimise the model further, we present a custom loss function for multi-label classification and a novel multi-channel architecture with differential training that gives the model the ability to use all input data efficiently. Our results show the proposed deep learning architecture's superior performance, resulting in a 13% improvement in AUC (Area Under the receiver operating characteristic Curve) over traditional models. We also demonstrate how to produce an importance ranking for the different data sources and the temporal relationships using a Shapley approach specific to these models. |
| Date: | 2021–11 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2111.09902&r= |
| By: | Tsang, Andrew |
| Abstract: | This paper applies causal machine learning methods to analyze the heterogeneous regional impacts of monetary policy in China. The method uncovers the heterogeneous regional im-pacts of different monetary policy stances on the provincial figures for real GDP growth, CPI inflation and loan growth compared to the national averages. The varying effects of expansionary and contractionary monetary policy phases on Chinese provinces are highlighted and explained. Subsequently, applying interpretable machine learning, the empirical results show that the credit channel is the main channel affecting the regional impacts of monetary policy. An imminent conclusion of the uneven provincial responses to the “one size fits all” monetary policy is that different policymakers should coordinate their efforts to search for the optimal fiscal and monetary policy mix. |
| Keywords: | China, monetary policy, regional heterogeneity, machine learning, shadow banking |
| JEL: | C54 C61 E52 R11 |
| Date: | 2021–07–28 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:110703&r= |
| By: | Edward Turner |
| Abstract: | Deep learning has shown remarkable results on Euclidean data (e.g. audio, images, text) however this type of data is limited in the amount of relational information it can hold. In mathematics we can model more general relational data in a graph structure while retaining Euclidean data as associated node or edge features. Due to the ubiquity of graph data, and its ability to hold multiple dimensions of information, graph deep learning has become a fast emerging field. We look at applying and optimising graph deep learning on a finance graph to produce more informed clusters of companies. Having clusters produced from multiple streams of data can be highly useful in quantitative finance; not only does it allow clusters to be tailored to the specific task but the culmination of multiple streams allows for cross source pattern recognition that would have otherwise gone unnoticed. This can provide financial institutions with an edge over competitors which is crucial in the heavily optimised world of trading. In this paper we use news co-occurrence and stock price for our data combination. We optimise our model to achieve an average testing precision of 78% and find a clear improvement in clustering capabilities when dual data sources are used; cluster purity rises from 32% for just vertex data and 42% for just edge data to 64% when both are used in comparisons to ground-truth Bloomberg clusters. The framework we provide utilises unsupervised learning which we view as key for future work due to the volume of unlabelled data in financial markets. |
| Date: | 2021–11 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2111.13519&r= |
| By: | Riccardo D'Adamo |
| Abstract: | This paper studies the problem of estimating individualized treatment rules when treatment effects are partially identified, as it is often the case with observational data. We first study the population problem of assigning treatment under partial identification and derive the population optimal policies using classic optimality criteria for decision under ambiguity. We then propose an algorithm for computation of the estimated optimal treatment policy and provide statistical guarantees for its convergence to the population counterpart. Our estimation procedure leverages recent advances in the orthogonal machine learning literature, while our theoretical results account for the presence of non-differentiabilities in the problem. The proposed methods are illustrated using data from the Job Partnership Training Act study. |
| Date: | 2021–11 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2111.10904&r= |
| By: | Beltempo, Marc (McGill University); Bresson, Georges (University of Paris 2); Lacroix, Guy (Université Laval) |
| Abstract: | Background: Adult studies have shown that nursing overtime and unit overcrowding is associated with increased adverse patient events but there exists little evidence for the Neonatal Intensive Care Unit (NICU). Objectives: To predict the onset on nosocomial infections and medical accidents in a NICU using machine learning models. Subjects: Retrospective study on the 7,438 neonates admitted in the CHU de Québec NICU (capacity of 51 beds) from 10 April 2008 to 28 March 2013. Daily administrative data on nursing overtime hours, total regular hours, number of admissions, patient characteristics, as well as information on nosocomial infections and on the timing and type of medical errors were retrieved from various hospital-level datasets. Methodology: We use a generalized mixed effects regression tree model (GMERT) to elaborate predictions trees for the two outcomes. Neonates' characteristics and daily exposure to numerous covariates are used in the model. GMERT is suitable for binary outcomes and is a recent extension of the standard tree-based method. The model allows to determine the most important predictors. Results: DRG severity level, regular hours of work, overtime, admission rates, birth weight and occupation rates are the main predictors for both outcomes. On the other hand, gestational age, C-Section, multiple births, medical/surgical and number of admissions are poor predictors. Conclusion: Prediction trees (predictors and split points) provide a useful management tool to prevent undesirable health outcomes in a NICU. |
| Keywords: | neonatal health outcomes, nursing overtime, machine learning, mixed effects regression tree |
| JEL: | I1 J2 C11 C14 C23 |
| Date: | 2020–03 |
| URL: | http://d.repec.org/n?u=RePEc:iza:izadps:dp13099&r= |
| By: | Martin Hackmann (UCLA, NBER, and CESifo); Joerg Heining (Institut für Arbeitsmarkt-und Berufsforschung (IAB)); Roman Klimke (Harvard University); Maria Polyakova (Stanford University, NBER and CESifo); Holger Seibert (Institut für Arbeitsmarkt-und Berufsforschung (IAB)) |
| Abstract: | Arrow (1963) hypothesized that demand-side moral hazard induced by health insurance leads to supply-side expansions in healthcare markets. Capturing these effects empirically has been challenging, as non-marginal insurance expansions are rare and detailed data on healthcare labor and capital is sparse. We combine administrative labor market data with the geographic variation in the rollout of a universal insurance program—the introduction of long-term care (LTC) insurance in Germany in 1995—to document a substantial expansion of the inpatient LTC labor market in response to insurance expansion. A 10 percentage point expansion in the share of insured elderly leads to 0.05 (7%) more inpatient LTC firms and four (13%) more workers per 1,000 elderly in Germany. Wages did not rise, but the quality of newly hired workers declined. We find suggestive evidence of a reduction in old-age mortality. Using a machine learning algorithm, we characterize counterfactual labor market biographies of potential inpatient LTC hires, finding that the reform moved workers into LTC jobs from unemployment and out of the labor force rather than from other sectors of the economy. We estimate that employing these additional workers in LTC is socially efficient if patients value the care provided by these workers at least at 25% of the market price for care. We show conceptually that, in the spirit of Harberger (1971), in a second-best equilibrium in which supply-side labor markets do not clear at perfectly competitive wages, subsidies for healthcare consumption along with the associated demand-side moral hazard can be welfare-enhancing. |
| Keywords: | long-term care, universal insurance expansion, Germany, LTC labor market, second-best efficiency |
| JEL: | D61 I11 I13 J21 J23 |
| Date: | 2021–11 |
| URL: | http://d.repec.org/n?u=RePEc:upj:weupjo:21-357&r= |
| By: | Alexandre Georgieff; Raphaela Hyee |
| Abstract: | Recent years have seen impressive advances in artificial intelligence (AI) and this has stoked renewed concern about the impact of technological progress on the labour market, including on worker displacement.This paper looks at the possible links between AI and employment in a cross-country context. It adapts the AI occupational impact measure developed by Felten, Raj and Seamans (2018[1]; 2019[2]) – an indicator measuring the degree to which occupations rely on abilities in which AI has made the most progress – and extends it to 23 OECD countries. The indicator, which allows for variations in AI exposure across occupations, as well as within occupations and across countries, is then matched to Labour Force Surveys, to analyse the relationship with employment.Over the period 2012-2019, employment grew in nearly all occupations analysed. Overall, there appears to be no clear relationship between AI exposure and employment growth. However, in occupations where computer use is high, greater exposure to AI is linked to higher employment growth. The paper also finds suggestive evidence of a negative relationship between AI exposure and growth in average hours worked among occupations where computer use is low.While further research is needed to identify the exact mechanisms driving these results, one possible explanation is that partial automation by AI increases productivity directly as well as by shifting the task composition of occupations towards higher value-added tasks. This increase in labour productivity and output counteracts the direct displacement effect of automation through AI for workers with good digital skills, who may find it easier to use AI effectively and shift to non-automatable, higher-value added tasks within their occupations. The opposite could be true for workers with poor digital skills, who may not be able to interact efficiently with AI and thus reap all potential benefits of the technology. |
| Keywords: | artificial intelligence, employment |
| JEL: | J21 J23 J24 O33 |
| Date: | 2021–12–15 |
| URL: | http://d.repec.org/n?u=RePEc:oec:elsaab:265-en&r= |