Motivated by the application to German interest rates, we propose a timevarying autoregressive model for short and long term prediction of time series that exhibit a temporary non-stationary behavior but are assumed to mean revert in the long run. We use a Bayesian formulation to incorporate prior assumptions on the mean reverting process in the model and thereby regularize predictions in the far future. We use MCMC-based inference by deriving relevant full conditional distributions and employ a Metropolis-Hastings within Gibbs Sampler approach to sample from the posterior (predictive) distribution. In combining data-driven short term predictions with long term distribution assumptions our model is competitive to the existing methods in the short horizon while yielding reasonable predictions in the long run. We apply our model to interest rate data and contrast the forecasting performance to the one of a 2-Additive-Factor Gaussian model as well as to the predictions of a dynamic Nelson-Siegel model.

This paper proposes a public-private insurance scheme for earthquakes and floods in Italy in which property-owners, the insurer and the government co-operate in risk financing. Our model departs from the existing literature by describing a public-private insurance intended to relieve the financial burden that natural events place on governments, while at the same time assisting individuals and protecting the insurance business. Hence, the business is aiming at maximizing social welfare rather than profits. Given the limited amount of data available on natural risks, expected losses per individual have been estimated through risk-modeling. In order to evaluate the insurer's loss profile, spatial correlation among insured assets has been evaluated by means of the Hoeffding bound for r-dependent random variables. Though earthquakes generate expected losses that are almost six times greater than floods, we found that the amount of public funds needed to manage the two perils is almost the same. We argue that this result is determined by a combination of the risk aversion of individuals and the shape of the loss distribution. Lastly, since earthquakes and floods are uncorrelated, we tested whether jointly managing the two perils can counteract the negative impact of spatial correlation. Some benefit from risk diversification emerged, though the probability of the government having to inject further capital might be considerable. Our findings suggest that, when not supported by the government, private insurance might either financially over-expose the insurer or set premiums so high that individuals would fail to purchase policies.

Urban systems, composed of households, businesses, and infrastructures, are continuously evolving and expanding. This has several implications because the impacts of disruptions, and the complexity and interdependence of systems, are rapidly increasing. Hence, we face a challenge in how to improve our understanding about the interdependencies among those entities, as well as their responses to disruptions. The aims of this study were to (1) create an agent that mimics the metabolism of a business or household that obtains supplies from and provides output to infrastructure systems; (2) implement a network of agents that exchange resources, as coordinated with a price mechanism; and (3) test the responses of this prototype model to disruptions. Our investigation resulted in the development of a business/household agent and a dynamically self-organizing mechanism of network coordination under disruption based on costs for production and transportation. Simulation experiments confirmed the feasibility of this new model for analyzing responses to disruptions. Among the nine disruption scenarios considered, in line with our expectations, the one combining the failures of infrastructure links and production processes had the most negative impact. We also identified areas for future research that focus on network topologies, mechanisms for resource allocation, and disruption generation.

We propose a multi-factor polynomial framework to model and hedge long-term electricity contracts with delivery period. This framework has several advantages: the computation of forwards, risk premium and correlation between different forwards are fully explicit, and the model can be calibrated to observed electricity forward curves easily and well. Electricity markets suffer from non-storability and poor medium- to long-term liquidity. Therefore, we suggest a rolling hedge which only uses liquid forward contracts and is risk-minimizing in the sense of F\"ollmer and Schweizer. We calibrate the model to over eight years of German power calendar year forward curves and investigate the quality of the risk-minimizing hedge over various time horizons.

This scientific research paper presents an innovative approach based on deep reinforcement learning (DRL) to solve the algorithmic trading problem of determining the optimal trading position at any point in time during a trading activity in stock markets. It proposes a novel DRL trading strategy so as to maximise the resulting Sharpe ratio performance indicator on a broad range of stock markets. Denominated the Trading Deep Q-Network algorithm (TDQN), this new trading strategy is inspired from the popular DQN algorithm and significantly adapted to the specific algorithmic trading problem at hand. The training of the resulting reinforcement learning (RL) agent is entirely based on the generation of artificial trajectories from a limited set of stock market historical data. In order to objectively assess the performance of trading strategies, the research paper also proposes a novel, more rigorous performance assessment methodology. Following this new performance assessment approach, promising results are reported for the TDQN strategy.

Retailers and major consumers of electricity generally purchase a critical percentage of their estimated electricity needs years ahead on the forward markets. This long-term electricity procurement task consists of determining when to buy electricity so that the resulting energy cost is minimised, and the forecast consumption is covered. In this scientific article, the focus is set on a yearly base load product, named calendar (CAL), which is tradable up to three years ahead of the delivery period. This research paper introduces a novel algorithm providing recommendations to either buy electricity now or wait for a future opportunity based on the history of CAL prices. This algorithm relies on deep learning forecasting techniques and on an indicator quantifying the deviation from a perfectly uniform reference procurement strategy. Basically, a new purchase operation is advised when this mathematical indicator hits the trigger associated with the market direction predicted by the forecaster. On average, the proposed approach surpasses benchmark procurement strategies and achieves a reduction in costs of 1.65% with respect to the perfectly uniform reference procurement strategy achieving the mean electricity price. Moreover, in addition to automating the electricity procurement task, this algorithm demonstrates more consistent results throughout the years compared to the benchmark strategies.

In this article we consider the infinite-horizon Merton investment-consumption problem in a constant-parameter Black - Scholes - Merton market for an agent with constant relative risk aversion R. The classical primal approach is to write down a candidate value function and to use a verification argument to prove that this is the solution to the problem. However, features of the problem take it outside the standard settings of stochastic control, and the existing primal verification proofs rely on parameter restrictions (especially, but not only, R<1), restrictions on the space of admissible strategies, or intricate approximation arguments.

The purpose of this paper is to show that these complications can be overcome using a simple and elegant argument involving a stochastic perturbation of the utility function.

We summarize the fundamental issues at stake in algorithmic trading, and the progress made in this field over the last twenty years. We first present the key problems of algorithmic trading, describing the concepts of optimal execution, optimal placement, and price impact. We then discuss the most recent advances in algorithmic trading through the use of Machine Learning, discussing the use of Deep Learning, Reinforcement Learning, and Generative Adversarial Networks.

We develop a model of bailout stigma where accepting a bailout signals a firm's balance-sheet weakness and worsens its funding prospect. To avoid stigma, a firm with high-quality legacy assets either withdraws from subsequent financing after receiving a bailout or refuses a bailout altogether to send a favorable signal. The former leads to a short-lived stimulation with subsequent market freeze even worse than if there were no bailout. The latter revives the funding market, albeit with delay, to the level achievable without any stigma. Strikingly, a bailout offer is most effective when many firms reject it (to build a favorable reputation) rather than accept it.

We analyze an optimal trade execution problem in a financial market with stochastic liquidity. To this end we set up a limit order book model in continuous time. Both order book depth and resilience are allowed to evolve randomly in time. We allow for trading in both directions and for c\`adl\`ag semimartingales as execution strategies. We derive a quadratic BSDE that under appropriate assumptions characterizes minimal execution costs and identify conditions under which an optimal execution strategy exists. We also investigate qualitative aspects of optimal strategies such as, e.g., appearance of strategies with infinite variation or existence of block trades and discuss connections with the discrete-time formulation of the problem. Our findings are illustrated in several examples.

Prior literature has argued that flood insurance maps may not capture the extent of flood risk. This paper performs a granular assessment of coastal flood risk in the mortgage market by using physical simulations of hurricane storm surge heights instead of using FEMA's flood insurance maps. Matching neighborhood-level predicted storm surge heights with mortgage files suggests that coastal flood risk may be large: originations and securitizations in storm surge areas have been rising sharply since 2012, while they remain stable when using flood insurance maps. Every year, more than 50 billion dollars of originations occur in storm surge areas outside of insurance floodplains. The share of agency mortgages increases in storm surge areas, yet remains stable in the flood insurance 100-year floodplain. Mortgages in storm surge areas are more likely to be complex: non-fully amortizing features such as interest-only or adjustable rates. Households may also be more vulnerable in storm surge areas: median household income is lower, the share of African Americans and Hispanics is substantially higher, the share of individuals with health coverage is lower. Price-to-rent ratios are declining in storm surge areas while they are increasing in flood insurance areas. This paper suggests that uncovering future financial flood risk requires scientific models that are independent of the flood insurance mapping process.

We provide one of the first systematic assessments of the development and determinants of economic anxiety at the onset of the coronavirus pandemic. Using a global dataset on internet searches and two representative surveys from the US, we document a substantial increase in economic anxiety during and after the arrival of the coronavirus. We also document a large dispersion in beliefs about the pandemic risk factors of the coronavirus, and demonstrate that these beliefs causally affect individuals' economic anxieties. Finally, we show that individuals' mental models of infectious disease spread understate non-linear growth and shape the extent of economic anxiety.

Forecasting stock prices can be interpreted as a time series prediction problem, for which Long Short Term Memory (LSTM) neural networks are often used due to their architecture specifically built to solve such problems. In this paper, we consider the design of a trading strategy that performs portfolio optimization using the LSTM stock price prediction for four different companies. We then customize the loss function used to train the LSTM to increase the profit earned. Moreover, we propose a data driven approach for optimal selection of window length and multi-step prediction length, and consider the addition of analyst calls as technical indicators to a multi-stack Bidirectional LSTM strengthened by the addition of Attention units. We find the LSTM model with the customized loss function to have an improved performance in the training bot over a regressive baseline such as ARIMA, while the addition of analyst call does improve the performance for certain datasets.

This paper considers how the effect of job displacement varies across different individuals. In particular, our interest centers on features of the distribution of the individual-level effect of job displacement. Identifying features of this distribution is particularly challenging -- e.g., even if we could randomly assign workers to be displaced or not, many of the parameters that we consider would not be point identified. We exploit our access to panel data, and our approach relies on comparing outcomes of displaced workers to outcomes the same workers would have experienced if they had not been displaced and if they maintained the same rank in the distribution of earnings as they had before they were displaced. Using data from the Displaced Workers Survey, we find that displaced workers earn about $157 per week less than they would have earned if they had not been displaced. We also find that there is substantial heterogeneity. We estimate that 42% of workers have higher earnings than they would have had if they had not been displaced and that a large fraction of workers have experienced substantially more negative effects than the average effect of displacement. Finally, we also document major differences in the distribution of the effect of job displacement across education levels, sex, age, and counterfactual earnings levels. Throughout the paper, we rely heavily on quantile regression. First, we use quantile regression as a flexible (yet feasible) first step estimator of conditional distributions and quantile functions that our main results build on. We also use quantile regression to study how covariates affect the distribution of the individual-level effect of job displacement.

Optimal order execution is widely studied by industry practitioners and academic researchers because it determines the profitability of investment decisions and high-level trading strategies, particularly those involving large volumes of orders. However, complex and unknown market dynamics pose enormous challenges for the development and validation of optimal execution strategies. We propose a model-free approach by training Reinforcement Learning (RL) agents in a realistic market simulation environment with multiple agents. First, we have configured a multi-agent historical order book simulation environment for execution tasks based on an Agent-Based Interactive Discrete Event Simulation (ABIDES) [arXiv:1904.12066]. Second, we formulated the problem of optimal execution in an RL setting in which an intelligent agent can make order execution and placement decisions based on market microstructure trading signals in HFT. Third, we developed and trained an RL execution agent using the Double Deep Q-Learning (DDQL) algorithm in the ABIDES environment. In some scenarios, our RL agent converges towards a Time-Weighted Average Price (TWAP) strategy. Finally, we evaluated the simulation with our RL agent by comparing the simulation on the actual market Limit Order Book (LOB) characteristics.

We analyze an optimal trade execution problem in a financial market with stochastic liquidity. To this end we set up a limit order book model in which both order book depth and resilience evolve randomly in time. Trading is allowed in both directions and at discrete points in time. We derive an explicit recursion that, under certain structural assumptions, characterizes minimal execution costs. We also discuss several qualitative aspects of optimal strategies, such as existence of profitable round trips or closing the position in one go, and compare our findings with the literature.

Natural hazards can considerably impact the overall society of a country. As some degree of public sector involvement is always necessary to deal with the consequences of natural disasters, central governments have increasingly invested in proactive risk management planning. In order to empower and involve the whole society, some countries have established public-private partnerships, mainly with the insurance industry, with satisfactorily outcomes. Although they have proven necessary and most often effective, the public-private initiatives have often incurred high debts or have failed to achieved the desired risk reduction objectives. We review the role of these partnerships in the management of natural risks, with particular attention to the insurance sector. Among other country-specific issues, poor risk knowledge and weak governance have widely challenged the initiatives during the recent years, while the future is threatened by the uncertainty of climate change and unsustainable development. In order to strengthen the country's resilience, a greater involvement of all segments of the community, especially the weakest layers, is needed and the management of natural risks should be included in a sustainable development plan.

We explain in a nontechnical fashion why dollar-neutral quant trading strategies, such as equities Statistical Arbitrage, suffered substantial losses (drawdowns) during the COVID-19 market selloff. We discuss: (i) why these strategies work during "normal" times; (ii) the market regimes when they work best; and (iii) their limitations and the reasons for why they "break" during extreme market events. An accompanying appendix (with a link to freely accessible source code) includes backtests for various strategies, which put flesh on and illustrate the discussion in the main text.

We consider a single-period portfolio selection problem for an investor, maximizing the expected ratio of the portfolio utility and the utility of a best asset taken in hindsight. The decision rules are based on the history of stock returns with unknown distribution. Assuming that the utility function is Lipschitz or H\"{o}lder continuous (the concavity is not required), we obtain high probability utility bounds under the sole assumption that the returns are independent and identically distributed. These bounds depend only on the utility function, the number of assets and the number of observations. For concave utilities similar bounds are obtained for the portfolios produced by the exponentiated gradient method. Also we use statistical experiments to study risk and generalization properties of empirically optimal portfolios. Herein we consider a model with one risky asset and a dataset, containing the stock prices from NYSE.

Digital contact tracing and analysis of social distancing from smartphone location data are two prime examples of non-therapeutic interventions used in many countries to mitigate the impact of the COVID-19 pandemic. While many understand the importance of trading personal privacy for the public good, others have been alarmed at the potential for surveillance via measures enabled through location tracking on smartphones. In our research, we analyzed massive yet atomic individual-level location data containing over 22 billion records from ten Blue (Democratic) and ten Red (Republican) cities in the U.S., based on which we present, herein, some of the first evidence of how Americans responded to the increasing concerns that government authorities, the private sector, and public health experts might use individual-level location data to track the COVID-19 spread. First, we found a significant decreasing trend of mobile-app location-sharing opt-out. Whereas areas with more Democrats were more privacy-concerned than areas with more Republicans before the advent of the COVID-19 pandemic, there was a significant decrease in the overall opt-out rates after COVID-19, and this effect was more salient among Democratic than Republican cities. Second, people who practiced social distancing (i.e., those who traveled less and interacted with fewer close contacts during the pandemic) were also less likely to opt-out, whereas the converse was true for people who practiced less social-distancing. This relationship also was more salient among Democratic than Republican cities. Third, high-income populations and males, compared with low-income populations and females, were more privacy-conscientious and more likely to opt-out of location tracking.

Increased testing for Covid-19 is seen as one of the most important steps to re-open the economy. The current paper considers Tennessee's "open-testing" policy where the state substantially increased testing while removing symptom requirements for individuals to be tested. To understand the effect of the policy, we combine standard identifying assumptions with additional weak assumptions to deal with non-random testing that lead to bounds on policy effects of interest. Our results suggest that Tennessee's open-testing policy has reduced total cases (which are not fully observed), confirmed cases, and trips to work among counties with a fast-growing number of confirmed cases.