|
on Market Microstructure |
| By: | Koti S. Jaddu; Paul A. Bilokon |
| Abstract: | High-frequency trading is prevalent, where automated decisions must be made quickly to take advantage of price imbalances and patterns in price action that forecast near-future movements. While many algorithms have been explored and tested, analytical methods fail to harness the whole nature of the market environment by focusing on a limited domain. With the evergrowing machine learning field, many large-scale end-to-end studies on raw data have been successfully employed to increase the domain scope for profitable trading but are very difficult to replicate. Combining deep learning on the order books with reinforcement learning is one way of breaking down large-scale end-to-end learning into more manageable and lightweight components for reproducibility, suitable for retail trading. The following work focuses on forecasting returns across multiple horizons using order flow imbalance and training three temporal-difference learning models for five financial instruments to provide trading signals. The instruments used are two foreign exchange pairs (GBPUSD and EURUSD), two indices (DE40 and FTSE100), and one commodity (XAUUSD). The performances of these 15 agents are evaluated through backtesting simulation, and successful models proceed through to forward testing on a retail trading platform. The results prove potential but require further minimal modifications for consistently profitable trading to fully handle retail trading costs, slippage, and spread fluctuation. |
| Date: | 2023–10 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2311.02088&r=mst |
| By: | Darrell Duffie; Michael Fleming; Frank Keane; Claire Nelson; Or Shachar; Peter Van Tassel |
| Abstract: | We show a significant loss in US Treasury market functionality when intensive use of dealer balance sheets is needed to intermediate bond markets, as in March 2020. Although yield volatility explains most of the variation in Treasury market liquidity over time, when dealer balance sheet utilization reaches sufficiently high levels, liquidity is much worse than predicted by yield volatility alone. This is consistent with the existence of occasionally binding constraints on the intermediation capacity of bond markets. |
| Keywords: | Treasury market, liquidity, volatility, dealer intermediation, value-at-risk |
| JEL: | G01 G1 G12 G18 E58 |
| Date: | 2023–10 |
| URL: | http://d.repec.org/n?u=RePEc:bis:biswps:1138&r=mst |
| By: | Simon Jurkatis; Andreas Schrimpf; Karamfil Todorov; Nicholas Vause |
| Abstract: | We find that clients with stronger past trading relationships with a dealer receive consistently better prices in corporate bond trading. The top 1% of relationship clients enjoy transaction costs that are 51% lower than those of the median client - an effect which was particularly beneficial when transaction costs spiked during the COVID-19 turmoil. We find clients' liquidity provision to be a key motive why dealers grant relationship discounts: clients to whom balance-sheet constrained dealers can turn as a source of liquidity are rewarded with relationship discounts. Another important motive for dealers to give discounts to relationship clients is because these clients generate the bulk of dealers' profits. Finally, we find no evidence that extraction of information from clients' order flow is related to relationship discounts. |
| Keywords: | corporate bonds, Covid-19, dealers, over-the-counter markets, trading relationships |
| JEL: | G12 G14 G23 G24 |
| Date: | 2023–11 |
| URL: | http://d.repec.org/n?u=RePEc:bis:biswps:1140&r=mst |
| By: | Peter Reinhard Hansen; Yiyao Luo |
| Abstract: | Time-varying volatility is an inherent feature of most economic time-series, which causes standard correlation estimators to be inconsistent. The quadrant correlation estimator is consistent but very inefficient. We propose a novel subsampled quadrant estimator that improves efficiency while preserving consistency and robustness. This estimator is particularly well-suited for high-frequency financial data and we apply it to a large panel of US stocks. Our empirical analysis sheds new light on intra-day fluctuations in market betas by decomposing them into time-varying correlations and relative volatility changes. Our results show that intraday variation in betas is primarily driven by intraday variation in correlations. |
| Date: | 2023–10 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2310.19992&r=mst |
| By: | Junqian Li (School of Economics, Shandong University); Yuqing Liu (School of Economics, University of Queensland, Brisbane, Australia); Nhan Buu Phan (School of Economics, University of Queensland, Brisbane, Australia); Shino Takayama (School of Economics, University of Queensland, Brisbane, Australia) |
| Abstract: | In this paper, we study the trading strategies of informed traders in a simulated asset market. There is a risky asset with two possible values, and participants receive private information about the value of the asset. Market maker’s quotes are computationally simulated. We study whether the trading behavior of informed traders—specifically, the frequency of manipulative trading versus honest trading—is influenced by various conditions, including the bid–ask spread, retrading possibilities, and the risk attitude of traders. Our findings suggest that manipulation occurs in both long (e.g., 15 periods) and short (e.g., five periods) trading rounds. Furthermore, there is a significant increase in the number of manipulators when the bid–ask spread is narrow rather than wide. Our results also indicate that risk-seeking participants engage in manipulation more frequently than other participants. |
| Date: | 2023–11 |
| URL: | http://d.repec.org/n?u=RePEc:qld:uq2004:665&r=mst |