Geometric Arbitrage Theory reformulates a generic asset model possibly allowing for arbitrage by packaging all assets and their forwards dynamics into a stochastic principal fibre bundle, with a connection whose parallel transport encodes discounting and portfolio rebalancing, and whose curvature measures, in this geometric language, the 'instantaneous arbitrage capability' generated by the market itself. The cashflow bundle is the vector bundle associated to this stochastic principal fibre bundle for the natural choice of the vector space fibre. The cashflow bundle carries a stochastic covariant differentiation induced by the connection on the principal fibre bundle. The link between arbitrage theory and spectral theory of the connection Laplacian on the vector bundle is given by the zero eigenspace resulting in a parametrization of all risk neutral measures equivalent to the statistical one. This indicates that a market satisfies the (NFLVR) condition if and only if $0$ is in the discrete spectrum of the connection Laplacian on the cash flow bundle or of the Dirac Laplacian of the twisted cash flow bundle with the exterior algebra bundle. We apply this result by extending Jarrow-Protter-Shimbo theory of asset bubbles for complete arbitrage free markets to markets not satisfying the (NFLVR). Moreover, by means of the Atiyah-Singer index theorem, we prove that the Euler characteristic of the asset nominal space is a topological obstruction to the the (NFLVR) condition, and, by means of the Bochner-Weitzenb\"ock formula, the non vanishing of the homology group of the cash flow bundle is revealed to be a topological obstruction to (NFLVR), too. Asset bubbles are defined, classified and decomposed for markets allowing arbitrage.

We present an artificial neural network (ANN) approach to value financial derivatives. Atypically to standard ANN applications, practitioners equally use option pricing models to validate market prices and to infer unobserved prices. Importantly, models need to generate realistic arbitrage-free prices, meaning that no option portfolio can lead to risk-free profits. The absence of arbitrage opportunities is guaranteed by penalizing the loss using soft constraints on an extended grid of input values. ANNs can be pre-trained by first calibrating a standard option pricing model, and then training an ANN to a larger synthetic dataset generated from the calibrated model. The parameters transfer as well as the non-arbitrage constraints appear to be particularly useful when only sparse or erroneous data are available. We also explore how deeper ANNs improve over shallower ones, as well as other properties of the network architecture. We benchmark our method against standard option pricing models, such as Heston with and without jumps. We validate our method both on training sets, and testing sets, namely, highlighting both their capacity to reproduce observed prices and predict new ones.

Ergodicity describes an equivalence between the expectation value and the time average of observables. Applied to human behaviour, ergodic theory reveals how individuals should tolerate risk in different environments. To optimise wealth over time, agents should adapt their utility function according to the dynamical setting they face. Linear utility is optimal for additive dynamics, whereas logarithmic utility is optimal for multiplicative dynamics. Whether humans approximate time optimal behavior across different dynamics is unknown. Here we compare the effects of additive versus multiplicative gamble dynamics on risky choice. We show that utility functions are modulated by gamble dynamics in ways not explained by prevailing economic theory. Instead, as predicted by time optimality, risk aversion increases under multiplicative dynamics, distributing close to the values that maximise the time average growth of wealth. We suggest that our findings motivate a need for explicitly grounding theories of decision-making on ergodic considerations.

We argue that a stochastic model of economic exchange, whose steady-state distribution is a Generalized Beta Prime (also known as GB2), and some unique properties of the latter, are the reason for GB2's success in describing wealth/income distributions. We use housing sale prices as a proxy to wealth/income distribution to numerically illustrate this point. We also explore parametric limits of the distribution to do so analytically. We discuss parametric properties of the inequality indices -- Gini, Hoover, Theil T and Theil L -- vis-a-vis those of GB2 and introduce a new inequality index, which serves a similar purpose. We argue that Hoover and Theil L are more appropriate measures for distributions with power-law dependencies, especially fat tails, such as GB2.

Pairs Trading is carried out in the financial market to earn huge profits from known equilibrium relation between pairs of stock. In financial markets, seldom it is seen that stock pairs are correlated at particular lead or lag. This lead-lag relationship has been empirically studied in various financial markets. Earlier research works have suggested various measures for identifying the best pairs for pairs trading, but they do not consider this lead-lag effect. The present study proposes a new distance measure which incorporates the lead-lag relationship between the stocks while selecting the best pairs for pairs trading. Further, the lead-lag value between the stocks is allowed to vary continuously over time. The proposed measures importance has been show-cased through experimentation on two different datasets, one corresponding to Indian companies and another corresponding to American companies. When the proposed measure is clubbed with SSD measure, i.e., when pairs are identified through optimising both these measures, then the selected pairs consistently generate the best profit, as compared to all other measures. Finally, possible generalisation and extension of the proposed distance measure have been discussed.

We advocate the use of Agnostic Allocation for the construction of long-only portfolios of stocks. We show that Agnostic Allocation Portfolios (AAPs) are a special member of a family of risk-based portfolios that are able to mitigate certain extreme features (excess concentration, high turnover, strong exposure to low-risk factors) of classical portfolio construction methods, while achieving similar performance. AAPs thus represent a very attractive alternative risk-based portfolio construction framework that can be implemented in different situations, with or without an active trading signal.

Existing research argues that countries increase their production baskets based on the available capabilities, adding products which require similar capabilities to those already produced, a process referred to as path dependency. Expansions to include goods that use divergent capabilities from those currently in the economy requires a structural change in available capabilities. Structural changes in existing capabilities contributes to countries economic growth and development. Economic development increases environmental risks from higher consumption of energy and natural resources. Managing that risk is critical and a transition to a green economy can help. The main objective of this research is to determine if structural changes or path dependency drives the expansion in production of green economy products. We consider a dataset with 138 countries over the period of 2008 to 2017, with a focus on specific case study examples, including all countries in the world and China. The results of this research show countries increased their green production baskets based on their available capabilities following path dependency as well as by expanding to products that path dependency does not predict. This suggests that, while path dependency may explain some expansion in green economies, additional theories are needed to fully explain observed green economic expansion.

Multi-agent learning is a promising method to simulate aggregate competitive behaviour in finance. Learning expert agents' reward functions through their external demonstrations is hence particularly relevant for subsequent design of realistic agent-based simulations. Inverse Reinforcement Learning (IRL) aims at acquiring such reward functions through inference, allowing to generalize the resulting policy to states not observed in the past. This paper investigates whether IRL can infer such rewards from agents within real financial stochastic environments: limit order books (LOB). We introduce a simple one-level LOB, where the interactions of a number of stochastic agents and an expert trading agent are modelled as a Markov decision process. We consider two cases for the expert's reward: either a simple linear function of state features; or a complex, more realistic non-linear function. Given the expert agent's demonstrations, we attempt to discover their strategy by modelling their latent reward function using linear and Gaussian process (GP) regressors from previous literature, and our own approach through Bayesian neural networks (BNN). While the three methods can learn the linear case, only the GP-based and our proposed BNN methods are able to discover the non-linear reward case. Our BNN IRL algorithm outperforms the other two approaches as the number of samples increases. These results illustrate that complex behaviours, induced by non-linear reward functions amid agent-based stochastic scenarios, can be deduced through inference, encouraging the use of inverse reinforcement learning for opponent-modelling in multi-agent systems.