This study first reconstructs three deep learning powered stock trading models and their associated strategies that are representative of distinct approaches to the problem and established upon different aspects of the many theories evolved around deep learning. It then seeks to compare the performance of these strategies from different perspectives through trading simulations ran on three scenarios when the benchmarks are kept at historical low points for extended periods of time. The results show that in extremely adverse market climates, investment portfolios managed by deep learning powered algorithms are able to avert accumulated losses by generating return sequences that shift the constantly negative CSI 300 benchmark return upward. Among the three, the LSTM model's strategy yields the best performance when the benchmark sustains continued loss.

Designing robust frameworks for precise prediction of future prices of stocks has always been considered a very challenging research problem. The advocates of the classical efficient market hypothesis affirm that it is impossible to accurately predict the future prices in an efficiently operating market due to the stochastic nature of the stock price variables. However, numerous propositions exist in the literature with varying degrees of sophistication and complexity that illustrate how algorithms and models can be designed for making efficient, accurate, and robust predictions of stock prices. We present a gamut of ten deep learning models of regression for precise and robust prediction of the future prices of the stock of a critical company in the auto sector of India. Using a very granular stock price collected at 5 minutes intervals, we train the models based on the records from 31st Dec, 2012 to 27th Dec, 2013. The testing of the models is done using records from 30th Dec, 2013 to 9th Jan 2015. We explain the design principles of the models and analyze the results of their performance based on accuracy in forecasting and speed of execution.

The phenomenon of brain drain, that is the emigration of highly skilled people, has many undesirable effects, particularly for developing countries. In this study, an agent-based model is developed to understand the dynamics of such emigration. We hypothesise that skilled people's emigration decisions are based on several factors including the overall economic and social difference between the home and host countries, people's ability and capacity to obtain good jobs and start a life abroad, and the barriers of moving abroad. Furthermore, the social network of individuals also plays a significant role. The model is validated using qualitative and quantitative pattern matching with real-world observations. Sensitivity and uncertainty analyses are performed in addition to several scenario analyses. Linear and random forest response surface models are created to provide quick predictions on the number of emigrants as well as to understand the effect sizes of individual parameters. Overall, the study provides an abstract model where brain drain dynamics can be explored. Findings from the simulation outputs show that future socioeconomic state of the country is more important than the current state, lack of barriers results in a large number of emigrants, and network effects ensue compounding effects on emigration. Upon further development and customisation, future versions can assist in the decision-making of social policymakers regarding brain drain.

This study derives the expected liquidity cost when performing the delta hedging process of a European option. This cost is represented by an integration formula that includes European option prices and a certain function depending on the delta process. We first define a unit liquidity cost and then show that the liquidity cost is a multiplication of the unit liquidity cost, stock price, supply curve parameter, and the square of the number of options. Using this formula, the expected liquidity cost before hedging can be calculated much faster than when using a Monte Carlo simulation. Numerically computed distributions of liquidity costs in special cases are also provided.

Bitcoin is a peer-to-peer electronic payment system that popularized rapidly in recent years. Usually, we need to query the complete history of Bitcoin blockchain data to acquire variables with economic meaning. This becomes increasingly difficult now with over 1.6 billion historical transactions on the Bitcoin blockchain. It is thus important to query Bitcoin transaction data in a way that is more efficient and provides economic insights. We apply cohort analysis that interprets Bitcoin blockchain data using methods developed for population data in social science. Specifically, we query and process the Bitcoin transaction input and output data within each daily cohort, which enables us to create datasets and visualizations for some key indicators of Bitcoin transactions, including the daily lifespan distributions of spent transaction output (STXO) and the daily age distributions of the accumulated unspent transaction output (UTXO). We provide a computationally feasible approach to characterize Bitcoin transactions, which paves the way for the future economic studies of Bitcoin.

University-industry research collaboration is one of the major research policy priorities of advanced economies. In this study, we try to identify the main drivers that could influence the propensity of academics to engage in research collaborations with the private sector, in order to better inform policies and initiatives to foster such collaborations. At this purpose, we apply an inferential model to a dataset of 32,792 Italian professors in order to analyze the relative impact of individual and contextual factors affecting the propensity of academics to engage in collaboration with industry, at overall level and across disciplines. The outcomes reveal that the typical profile of the professor collaborating with industry is a male under age 40, full professor, very high performer, with highly diversified research, and who has a certain tradition in collaborating with industry. This professor is likely to be part of a staff used to collaborating with industry, in a small university, typically a polytechnic, located in the north of the country.

Production networks are integral to economic dynamics, yet dis-aggregated network data on inter-firm trade is rarely collected and often proprietary. Here we situate company-level production networks among networks from other domains according to their local connectivity structure. Through this lens, we study a regional and a national network of inferred trade relationships reconstructed from Dutch national economic statistics and re-interpret prior empirical findings. We find that company-level production networks have so-called functional structure, as previously identified in protein-protein interaction (PPI) networks. Functional networks are distinctive in their over-representation of closed squares, which we quantify using an existing measure called spectral bipartivity. Shared local connectivity structure lets us ferry insights between domains. PPI networks are shaped by complementarity, rather than homophily, and we use multi-layer directed configuration models to show that this principle explains the emergence of functional structure in production networks. Companies are especially similar to their close competitors, not to their trading partners. Our findings have practical implications for the analysis of production networks and a thorough understanding of their local connectivity structure will help us better reason about the micro-economic mechanisms behind their routine function, failure, and growth.

We document the evolution of labour market power by employers on the US and Peruvian labour markets during the 2010s. Making use of a structural estimation model of labour market dynamics, we estimate differences in market power that workers face depending on their sector of activity, their age, sex, location and educational level. In particular, we show that differences in cross-sectional market power are significant and higher than variations over the ten-year time span of our data. In contrast to findings of labour market power in developed countries such as the US, we document significant market power of employers in Peru vis-\`a-vis the tertiary educated workforce, regardless of age and sector. In contrast, for the primary educated workforce, market power seems to be high in (private) services and manufacturing. For secondary educated workers, only the mining sector stands out as moderately more monopsonistic than the rest of the labour market. We also show that at least for the 2010s, labour market power declined in Peru. We contrast these findings with similar estimates obtained for the United States where we are able to show that increases in labour market power are particularly acute in certain sectors, such as agriculture and entertainment and recreational services, as well as in specific geographic areas where these sectors are dominant. Moreover, we show that for the US, the labour market power has gradually increased over the past ten years, in line with the general rise in inequality. Importantly, we show that the pervasive gender pay gap cannot be linked to differential market power as men face higher labour market power than women. We also discuss possible reasons for these findings, including for the differences of labour market power across skill levels. Especially, we discuss the reasons for polarization of market power that are specific to sectors and locations.

We develop a novel Monte Carlo algorithm for the vector consisting of the supremum, the time at which the supremum is attained and the position of an exponentially tempered L\'{e}vy process. The algorithm, based on the increments of the process without tempering, converges geometrically fast (as a function of the computational cost) for discontinuous and locally Lipschitz functions of the vector. We prove that the corresponding multilevel Monte Carlo estimator has optimal computational complexity (i.e. of order $\epsilon^{-2}$ if the mean squared error is at most $\epsilon^{2}$) and provide its central limit theorem (CLT). Using the CLT we construct confidence intervals for barrier option prices and various risk measures based on drawdown under the tempered stable (CGMY) model calibrated/estimated on real-world data. We provide non-asymptotic and asymptotic comparisons of our algorithm with existing approximations, leading to rule-of-thumb guidelines for users to the best method for a given set of parameters, and illustrate its performance with numerical examples.

An unconventional approach for optimal stopping under model ambiguity is introduced. Besides ambiguity itself, we take into account how ambiguity-averse an agent is. This inclusion of ambiguity attitude, via an $\alpha$-maxmin nonlinear expectation, renders the stopping problem time-inconsistent. We look for subgame perfect equilibrium stopping policies, formulated as fixed points of an operator. For a one-dimensional diffusion with drift and volatility uncertainty, we show that every equilibrium can be obtained through a fixed-point iteration. This allows us to capture much more diverse behavior, depending on an agent's ambiguity attitude, beyond the standard worst-case (or best-case) analysis. In a concrete example of real options valuation under volatility uncertainty, all equilibrium stopping policies, as well as the best one among them, are fully characterized. It demonstrates explicitly the effect of ambiguity attitude on decision making: the more ambiguity-averse, the more eager to stop -- so as to withdraw from the uncertain environment. The main result hinges on a delicate analysis of continuous sample paths in the canonical space and the capacity theory. To resolve measurability issues, a generalized measurable projection theorem, new to the literature, is also established.

Portfolio optimization approaches inevitably rely on multivariate modeling of markets and the economy. In this paper, we address three sources of error related to the modeling of these complex systems: 1. oversimplifying hypothesis; 2. uncertainties resulting from parameters' sampling error; 3. intrinsic non-stationarity of these systems. For what concerns point 1. we propose a L0-norm sparse elliptical modeling and show that sparsification is effective. The effects of points 2. and 3. are quantifified by studying the models' likelihood in- and out-of-sample for parameters estimated over train sets of different lengths. We show that models with larger off-sample likelihoods lead to better performing portfolios up to when two to three years of daily observations are included in the train set. For larger train sets, we found that portfolio performances deteriorate and detach from the models' likelihood, highlighting the role of non-stationarity. We further investigate this phenomenon by studying the out-of-sample likelihood of individual observations showing that the system changes significantly through time. Larger estimation windows lead to stable likelihood in the long run, but at the cost of lower likelihood in the short-term: the `optimal' fit in finance needs to be defined in terms of the holding period. Lastly, we show that sparse models outperform full-models in that they deliver higher out of sample likelihood, lower realized portfolio volatility and improved portfolios' stability, avoiding typical pitfalls of the Mean-Variance optimization.

This paper aims to further understand the main factors influencing the behavioural intentions (BI) of private vehicle users towards public transport to provide policymakers and public transport operators with the tools they need to attract more private vehicle users. As service quality, satisfaction and attitudes towards public transport are considered the main motivational forces behind the BI of public transport users, this research analyses 26 indicators frequently associated with these constructs for both public transport users and private vehicle users. Non-parametric tests and ordinal logit models have been applied to an online survey asked in Madrid's metropolitan area with a sample size of 1,025 respondents (525 regular public transport users and 500 regular private vehicle users). In order to achieve a comprehensive analysis and to deal with heterogeneity in perceptions, 338 models have been developed for the entire sample and for 12 users' segments. The results led to the identification of indicators with no significant differences between public transport and private vehicle users in any of the segments being considered (punctuality, information and low-income), as well as those that did show significant differences in all the segments (proximity, intermodality, save time and money, and lifestyle). The main differences between public transport and private vehicle users were found in the attitudes towards public transport and for certain user segments (residents in the city centre, males, young, with university qualification and with incomes above 2,700EUR/month). Findings from this study can be used to develop policies and recommendations for persuading more private vehicle users to use the public transport services.

This study on public-private research collaboration measures the variation over time of the propensity of academics to collaborate with colleagues from private companies. It also investigates the change in weights of the main drivers underlying the academics' propensity to collaborate, and whether the type profile of the collaborating academics changes. To do this, the study applies an inferential model on a dataset of professors working in Italian universities in consecutive periods, 2010-2013 and 2014-2017. The results, obtained at overall and field levels, support the formulation of policies aimed at fostering public-private research collaborations, and should be taken into account in post-assessment of their effectiveness.

We present a method for constructing Constant Function Market Makers (CFMMs) whose portfolio value functions match a desired payoff. More specifically, we show that the space of concave, nonnegative, nondecreasing, 1-homogeneous payoff functions and the space of convex CFMMs are equivalent; in other words, every CFMM has a concave, nonnegative, nondecreasing, 1-homogeneous payoff function, and every payoff function with these properties has a corresponding convex CFMM. We demonstrate a simple method for recovering a CFMM trading function that produces this desired payoff. This method uses only basic tools from convex analysis and is intimately related to Fenchel conjugacy. We demonstrate our result by constructing trading functions corresponding to basic payoffs, as well as standard financial derivatives such as options and swaps.

This paper presents an optimal strategy for portfolio liquidation under discrete time conditions. We assume that N risky assets held will be liquidated according to the same time interval and order quantity, and the basic price processes of assets are generated by an N-dimensional independent standard Brownian motion. The permanent impact generated by an asset in the portfolio during the liquidation will affect all assets, and the temporary impact generated by one asset will only affect itself. On this basis, we establish a liquidation cost model based on the VaR measurement and obtain an optimal liquidation time under discrete-time conditions. The optimal solution shows that the liquidation time is only related to the temporary impact rather than the permanent impact. In the simulation analysis, we give the relationship between volatility parameters, temporary price impact and the optimal liquidation strategy.

We develop a computational method for expected functionals of the drawdown and its duration in exponential L\'evy models. It is based on a novel simulation algorithm for the joint law of the state, supremum and time the supremum is attained of the Gaussian approximation of a general L\'evy process. We bound the bias for various locally Lipschitz and discontinuous payoffs arising in applications and analyse the computational complexities of the corresponding Monte Carlo and multilevel Monte Carlo estimators. Monte Carlo methods for L\'evy processes (using Gaussian approximation) have been analysed for Lipschitz payoffs, in which case the computational complexity of our algorithm is up to two orders of magnitude smaller when the jump activity is high. At the core of our approach are bounds on certain Wasserstein distances, obtained via the novel SBG coupling between a L\'evy process and its Gaussian approximation. Numerical performance, based on the implementation in the dedicated GitHub repository, exhibits a good agreement with our theoretical bounds.

In this paper monetary risk measures that are positively superhomogeneous, called star-shaped risk measures, are characterized and their properties studied. The measures in this class, which arise when the controversial subadditivity property of coherent risk measures is dispensed with and positive homogeneity is weakened, include all practically used risk measures, in particular, both convex risk measures and Value-at-Risk.

From a financial viewpoint, our relaxation of convexity is necessary to quantify the capital requirements for risk exposure in the presence of competitive delegation or robust aggregation mechanisms.

From a decision theoretical perspective, star-shaped risk measures emerge from variational preferences when risk mitigation strategies can be adopted by a rational decision maker.