arXiv subjects

Tim Gebbie

Publications and source records attributed to Tim Gebbie.

At least 37 records · Page 2Linked to original sources

arXiv2024

Geometric insights into robust portfolio construction

We investigate and extend the result that an alpha-weight angle from unconstrained quadratic portfolio optimisations has an upper bound dependent on the condition number of the covariance matrix. This is known to imply that better conditioned covariance matrices produce weights from unconstrained mean-variance optimisations that are better aligned with each assets expected return. Here we relate the inequality between the alpha-weight angle and the condition number to extend the result to include portfolio optimisations with gearing constraints to provide an extended family of robust optimisations. We use this to argue that in general the equally weighted portfolio is not preferable to the mean-variance portfolio even with poor forecast ability and a badly conditioned covariance matrix. We confirm the distribution free theoretical arguments with a simple Gaussian simulation.

q-fin.PM↗

arXiv2024

Representation Learning for Regime detection in Block Hierarchical Financial Markets

We consider financial market regime detection from the perspective of deep representation learning of the causal information geometry underpinning traded asset systems using a hierarchical correlation structure to characterise market evolution. We assess the robustness of three toy models: SPDNet, SPD-NetBN and U-SPDNet whose architectures respect the underlying Riemannian manifold of input block hierarchical SPD correlation matrices. Market phase detection for each model is carried out using three data configurations: randomised JSE Top 60 data, synthetically-generated block hierarchical SPD matrices and block-resampled chronology-preserving JSE Top 60 data. We show that using a singular performance metric is misleading in our financial market investment use cases where deep learning models overfit in learning spatio-temporal correlation dynamics.

q-fin.ST↗

arXiv2024

Many learning agents interacting with an agent-based market model

We consider the dynamics and the interactions of multiple reinforcement learning optimal execution trading agents interacting with a reactive Agent-Based Model (ABM) of a financial market in event time. The model represents a market ecology with 3-trophic levels represented by: optimal execution learning agents, minimally intelligent liquidity takers, and fast electronic liquidity providers. The optimal execution agent classes include buying and selling agents that can either use a combination of limit orders and market orders, or only trade using market orders. The reward function explicitly balances trade execution slippage against the penalty of not executing the order timeously. This work demonstrates how multiple competing learning agents impact a minimally intelligent market simulation as functions of the number of agents, the size of agents' initial orders, and the state spaces used for learning. We use phase space plots to examine the dynamics of the ABM, when various specifications of learning agents are included. Further, we examine whether the inclusion of optimal execution agents that can learn is able to produce dynamics with the same complexity as empirical data. We find that the inclusion of optimal execution agents changes the stylised facts produced by ABM to conform more with empirical data, and are a necessary inclusion for ABMs investigating market micro-structure. However, including execution agents to chartist-fundamentalist-noise ABMs is insufficient to recover the complexity observed in empirical data.

q-fin.TR↗

arXiv2024

Anomalous diffusion and price impact in the fluid-limit of an order book

We extend a Discrete Time Random Walk (DTRW) numerical scheme to simulate the anomalous diffusion of financial market orders in a simulated order book. Here using random walks with Sibuya waiting times to include a time-dependent stochastic forcing function with non-uniformly sampled times between order book events in the setting of fractional diffusion. This models the fluid limit of an order book by modelling the continuous arrival, cancellation and diffusion of orders in the presence of information shocks. We study the impulse response and stylised facts of orders undergoing anomalous diffusion for different forcing functions and model parameters. Concretely, we demonstrate the price impact for flash limit-orders and market orders and show how the numerical method generate kinks in the price impact. We use cubic spline interpolation to generate smoothed price impact curves. The work promotes the use of non-uniform sampling in the presence of diffusive dynamics as the preferred simulation method.

q-fin.CP↗

arXiv2023

A simple learning agent interacting with an agent-based market model

We consider the learning dynamics of a single reinforcement learning optimal execution trading agent when it interacts with an event driven agent-based financial market model. Trading takes place asynchronously through a matching engine in event time. The optimal execution agent is considered at different levels of initial order-sizes and differently sized state spaces. The resulting impact on the agent-based model and market are considered using a calibration approach that explores changes in the empirical stylised facts and price impact curves. Convergence, volume trajectory and action trace plots are used to visualise the learning dynamics. Here the smaller state space agents had the number of states they visited converge much faster than the larger state space agents, and they were able to start learning to trade intuitively using the spread and volume states. We find that the moments of the model are robust to the impact of the learning agents except for the Hurst exponent, which was lowered by the introduction of strategic order-splitting. The introduction of the learning agent preserves the shape of the price impact curves but can reduce the trade-sign auto-correlations when their trading volumes increase.

q-fin.TR↗

arXiv2023

Variations on the Reinforcement Learning performance of Blackjack

Blackjack or "21" is a popular card-based game of chance and skill. The objective of the game is to win by obtaining a hand total higher than the dealer's without exceeding 21. The ideal blackjack strategy will maximize financial return in the long run while avoiding gambler's ruin. The stochastic environment and inherent reward structure of blackjack presents an appealing problem to better understand reinforcement learning agents in the presence of environment variations. Here we consider a q-learning solution for optimal play and investigate the rate of learning convergence of the algorithm as a function of deck size. A blackjack simulator allowing for universal blackjack rules is also implemented to demonstrate the extent to which a card counter perfectly using the basic strategy and hi-lo system can bring the house to bankruptcy and how environment variations impact this outcome. The novelty of our work is to place this conceptual understanding of the impact of deck size in the context of learning agent convergence.

cs.AI↗

arXiv2021

Agglomerative Likelihood Clustering

We consider the problem of fast time-series data clustering. Building on previous work modeling the correlation-based Hamiltonian of spin variables we present an updated fast non-expensive Agglomerative Likelihood Clustering algorithm (ALC). The method replaces the optimized genetic algorithm based approach (f-SPC) with an agglomerative recursive merging framework inspired by previous work in Econophysics and Community Detection. The method is tested on noisy synthetic correlated time-series data-sets with built-in cluster structure to demonstrate that the algorithm produces meaningful non-trivial results. We apply it to time-series data-sets as large as 20,000 assets and we argue that ALC can reduce compute time costs and resource usage cost for large scale clustering for time-series applications while being serialized, and hence has no obvious parallelization requirement. The algorithm can be an effective choice for state-detection for online learning in a fast non-linear data environment because the algorithm requires no prior information about the number of clusters.

q-fin.CP↗

arXiv2021

Simulation and estimation of an agent-based market-model with a matching engine

An agent-based model with interacting low frequency liquidity takers inter-mediated by high-frequency liquidity providers acting collectively as market makers can be used to provide realistic simulated price impact curves. This is possible when agent-based model interactions occur asynchronously via order matching using a matching engine in event time to replace sequential calendar time market clearing. Here the matching engine infrastructure has been modified to provide a continuous feed of order confirmations and updates as message streams in order to conform more closely to live trading environments. The resulting trade and quote message data from the simulations are then aggregated, calibrated and visualised. Various stylised facts are presented along with event visualisations and price impact curves. We argue that additional realism in modelling can be achieved with a small set of agent parameters and simple interaction rules once interactions are reactive, asynchronous and in event time. We argue that the reactive nature of market agents may be a fundamental property of financial markets and when accounted for can allow for parsimonious modelling without recourse to additional sources of noise.

q-fin.TR↗

arXiv2021

Simulation and estimation of a point-process market-model with a matching engine

The extent to which a matching engine can cloud the modelling of underlying order submission and management processes in a financial market remains an unanswered concern with regards to market models. Here we consider a 10-variate Hawkes process with simple rules to simulate common order types which are submitted to a matching engine. Hawkes processes can be used to model the time and order of events, and how these events relate to each other. However, they provide a freedom with regards to implementation mechanics relating to the prices and volumes of injected orders. This allows us to consider a reference Hawkes model and two additional models which have rules that change the behaviour of limit orders. The resulting trade and quote data from the simulations are then calibrated and compared with the original order generating process to determine the extent with which implementation rules can distort model parameters. Evidence from validation and hypothesis tests suggest that the true model specification can be significantly distorted by market mechanics, and that practical considerations not directly due to model specification can be important with regards to model identification within an inherently asynchronous trading environment.

q-fin.TR↗

arXiv2021

The Epps effect under alternative sampling schemes

Time and the choice of measurement time scales is fundamental to how we choose to represent information and data in finance. This choice implies both the units and the aggregation scales for the resulting statistical measurables used to describe a financial system. It also defines how we measure the relationship between different traded instruments. As we move from high-frequency time scales, when individual trade and quote events occur, to the mesoscales when correlations emerge in ways that can conform to various latent models; it remains unclear what choice of time and sampling rates are appropriate to faithfully capture system dynamics and asset correlations for decision making. The Epps effect is the key phenomenology that couples the emergence of correlations to the choice of sampling time scales. Here we consider and compare the Epps effect under different sampling schemes in order to contrast three choices of time: calendar time, volume time and trade time. Using a toy model based on a Hawkes process, we are able to achieve simulation results that conform well with empirical dynamics. Concretely, we find that the Epps effect is present under all three definitions of time and that correlations emerge faster under trade time compared to calendar time, whereas correlations emerge linearly under volume time.

q-fin.ST↗

arXiv2021

The efficient frontiers of mean-variance portfolio rules under distribution misspecification

Mean-variance portfolio decisions that combine prediction and optimisation have been shown to have poor empirical performance. Here, we consider the performance of various shrinkage methods by their efficient frontiers under different distributional assumptions to study the impact of reasonable departures from Normality. Namely, we investigate the impact of first-order auto-correlation, second-order auto-correlation, skewness, and excess kurtosis. We show that the shrinkage methods tend to re-scale the sample efficient frontier, which can change based on the nature of local perturbations from Normality. This re-scaling implies that the standard approach of comparing decision rules for a fixed level of risk aversion is problematic, and more so in a dynamic market setting. Our results suggest that comparing efficient frontiers has serious implications which oppose the prevailing thinking in the literature. Namely, that sample estimators out-perform Stein type estimators of the mean, and that improving the prediction of the covariance has greater importance than improving that of the means.

q-fin.PM↗

arXiv2021

Calibrating an adaptive Farmer-Joshi agent-based model for financial markets

We replicate the contested calibration of the Farmer and Joshi agent based model of financial markets using a genetic algorithm and a Nelder-Mead with threshold accepting algorithm following Fabretti. The novelty of the Farmer-Joshi model is that the dynamics are driven by trade entry and exit thresholds alone. We recover the known claim that some important stylized facts observed in financial markets cannot be easily found under calibration -- in particular those relating to the auto-correlations in the absolute values of the price fluctuations, and sufficient kurtosis. However, rather than concerns relating to the calibration method, what is novel here is that we extended the Farmer-Joshi model to include agent adaptation using an Brock and Hommes approach to strategy fitness based on trading strategy profitability. We call this an adaptive Farmer-Joshi model: the model allows trading agents to switch between strategies by favouring strategies that have been more profitable over some period of time determined by a free-parameter fixing the profit monitoring time-horizon. In the adaptive model we are able to calibrate and recover additional stylized facts, despite apparent degeneracy's. This is achieved by combining the interactions of trade entry levels with trade strategy switching. We use this to argue that for low-frequency trading across days, as calibrated to daily sampled data, feed-backs can be accounted for by strategy die-out based on intermediate term profitability; we find that the average trade monitoring horizon is approximately two to three months (or 40 to 60 days) of trading.

q-fin.CP↗

arXiv2021

CoinTossX: An open-source low-latency high-throughput matching engine

We deploy and demonstrate the CoinTossX low-latency, high-throughput, open-source matching engine with orders sent using the Julia and Python languages. We show how this can be deployed for small-scale local desk-top testing and discuss a larger scale, but local hosting, with multiple traded instruments managed concurrently and managed by multiple clients. We then demonstrate a cloud based deployment using Microsoft Azure, with large-scale industrial and simulation research use cases in mind. The system is exposed and interacted with via sockets using UDP SBE message protocols and can be monitored using a simple web browser interface using HTTP. We give examples showing how orders can be be sent to the system and market data feeds monitored using the Julia and Python languages. The system is developed in Java with orders submitted as binary encodings (SBE) via UDP protocols using the Aeron Media Driver as the low-latency, high throughput message transport. The system separates the order-generation and simulation environments e.g. agent-based model simulation, from the matching of orders, data-feeds and various modularised components of the order-book system. This ensures a more natural and realistic asynchronicity between events generating orders, and the events associated with order-book dynamics and market data-feeds. We promote the use of Julia as the preferred order submission and simulation environment.

cs.DC↗

arXiv2020

Malliavin-Mancino estimators implemented with non-uniform fast Fourier transforms

We implement and test kernel averaging Non-Uniform Fast Fourier Transform (NUFFT) methods to enhance the performance of correlation and covariance estimation on asynchronously sampled event-data using the Malliavin-Mancino Fourier estimator. The methods are benchmarked for Dirichlet and Fejér Fourier basis kernels. We consider test cases formed from Geometric Brownian motions to replicate synchronous and asynchronous data for benchmarking purposes. We consider three standard averaging kernels to convolve the event-data for synchronisation via over-sampling for use with the Fast Fourier Transform (FFT): the Gaussian kernel, the Kaiser-Bessel kernel, and the exponential of semi-circle kernel. First, this allows us to demonstrate the performance of the estimator with different combinations of basis kernels and averaging kernels. Second, we investigate and compare the impact of the averaging scales explicit in each averaging kernel and its relationship between the time-scale averaging implicit in the Malliavin-Mancino estimator. Third, we demonstrate the relationship between time-scale averaging based on the number of Fourier coefficients used in the estimator to a theoretical model of the Epps effect. We briefly demonstrate the methods on Trade-and-Quote (TAQ) data from the Johannesburg Stock Exchange to make an initial visualisation of the correlation dynamics for various time-scales under market microstructure.

q-fin.CP↗

arXiv2020

Comparing the market microstructure between two South African exchanges

We consider shared listings on two South African equity exchanges: the Johannesburg Stock Exchange (JSE) and the A2X Exchange. A2X is an alternative exchange that provides for both shared listings and new listings within the financial market ecosystem of South Africa. From a science perspective it provides the opportunity to compare markets trading similar shares, in a similar regulatory and economic environment, but with vastly different liquidity, costs and business models. A2X currently has competitive settlement and transaction pricing when compared to the JSE, but the JSE has deeper liquidity. In pursuit of an empirical understanding of how these differences relate to their respective price response dynamics, we compare the distributions and auto-correlations of returns on different time scales; we compare price impact and master curves; and we compare the cost of trading on each exchange. This allows us to empirically compare the two markets. We find that various stylised facts become similar as the measurement or sampling time scale increase. However, the same securities can have vastly different price responses irrespective of time scales. This is not surprising given the different liquidity and order-book resilience. Here we demonstrate that direct costs dominate the cost of trading, and the importance of competitively positioning cost ceilings. Universality is crucial for being able to meaningfully compare cross-exchange price responses, but in the case of A2X, it has yet to emerge in a meaningful way due to the infancy of the exchange -- making meaningful comparisons difficult.

q-fin.ST↗

arXiv2020

Learning low-frequency temporal patterns for quantitative trading

We consider the viability of a modularised mechanistic online machine learning framework to learn signals in low-frequency financial time series data. The framework is proved on daily sampled closing time-series data from JSE equity markets. The input patterns are vectors of pre-processed sequences of daily, weekly and monthly or quarterly sampled feature changes. The data processing is split into a batch processed step where features are learnt using a stacked autoencoder via unsupervised learning, and then both batch and online supervised learning are carried out using these learnt features, with the output being a point prediction of measured time-series feature fluctuations. Weight initializations are implemented with restricted Boltzmann machine pre-training, and variance based initializations. Historical simulations are then run using an online feedforward neural network initialised with the weights from the batch training and validation step. The validity of results are considered under a rigorous assessment of backtest overfitting using both combinatorially symmetrical cross validation and probabilistic and deflated Sharpe ratios. Results are used to develop a view on the phenomenology of financial markets and the value of complex historical data-analysis for trading under the unstable adaptive dynamics that characterise financial markets.

q-fin.ST↗

arXiv2020

A Framework for Online Investment Algorithms

The artificial segmentation of an investment management process into a workflow with silos of offline human operators can restrict silos from collectively and adaptively pursuing a unified optimal investment goal. To meet the investor's objectives, an online algorithm can provide an explicit incremental approach that makes sequential updates as data arrives at the process level. This is in stark contrast to offline (or batch) processes that are focused on making component level decisions prior to process level integration. Here we present and report results for an integrated, and online framework for algorithmic portfolio management. This article provides a workflow that can in-turn be embedded into a process level learning framework. The workflow can be enhanced to refine signal generation and asset-class evolution and definitions. Our results confirm that we can use our framework in conjunction with resampling methods to outperform naive market capitalisation benchmarks while making clear the extent of back-test over-fitting. We consider such an online update framework to be a crucial step towards developing intelligent portfolio selection algorithms that integrate financial theory, investor views, and data analysis with process-level learning.

q-fin.PM↗

arXiv2020

Revisiting the Epps effect using volume time averaging: An exercise in R

We revisit and demonstrate the Epps effect using two well-known non-parametric covariance estimators; the Malliavin and Mancino (MM), and Hayashi and Yoshida (HY) estimators. We show the existence of the Epps effect in the top 10 stocks from the Johannesburg Stock Exchange (JSE) by various methods of aggregating Trade and Quote (TAQ) data. Concretely, we compare calendar time sampling with two volume time sampling methods: asset intrinsic volume time averaging, and volume time averaging synchronised in volume time across assets relative to the least and most liquid asset clocks. We reaffirm the argument made in much of the literature that the MM estimator is more representative of trade time reality because it does not over-estimate short-term correlations in an asynchronous event driven world. We confirm well known market phenomenology with the aim of providing some standardised R based simulation tools.

q-fin.CP↗