|
on Market Microstructure |
| By: | Mengoli, Stefano; Pagano, Marco; Pattitoni, Pierpaolo |
| Abstract: | Retail investors pay over twice as much attention to local companies than non-local ones, based on Google searches. News volume and volatility amplify this attention gap. Attention appears causally related to perceived proximity: first, acquisition by a nonlocal company is associated with less attention by locals, and more by nonlocals close to the acquirer; second, COVID-19 travel restrictions correlate with a drop in relative attention to nonlocal companies, especially in locations with fewer flights after the outbreak. Finally, local attention predicts volatility, bid-ask spreads and nonlocal attention, not viceversa. These findings are consistent with local investors having an information-processing advantage. |
| Keywords: | attention,retail investors,local investors,distance,news,liquidity,volatility |
| JEL: | D83 G11 G12 G14 G50 L86 R32 |
| Date: | 2021 |
| URL: | http://d.repec.org/n?u=RePEc:zbw:cfswop:671&r= |
| By: | Rama Cont; Mihai Cucuringu; Chao Zhang |
| Abstract: | We quantify the impact of order flow imbalances (OFIs) on price movements in equity markets. First, we examine the contemporaneous \textbf{price impact} of \textit{multi-level OFIs} and reveal their additional explanatory power. We propose an \textit{integrated OFI} by applying PCA to OFIs across different levels, leading to superior results in both in-sample and out-of-sample. Second, we introduce a method to examine the existence of \textbf{cross-impact}. Unlike previous works, we leverage LASSO to highlight the sparsity of the cross-impact terms. Compared with the best-level OFIs price-impact model, the cross-impact model with the entire cross-sectional best-level OFIs provides additional explanatory power. However, once multi-level OFIs have been incorporated, cross-impact terms cannot provide additional explanatory power. Last, we apply price-impact and cross-impact models to predict \textbf{future returns}, and provide evidence that cross-sectional OFIs significantly increase both in-sample and out-of-sample $R^2$. |
| Date: | 2021–12 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2112.13213&r= |
| By: | Kieran Wood; Sven Giegerich; Stephen Roberts; Stefan Zohren |
| Abstract: | Deep learning architectures, specifically Deep Momentum Networks (DMNs) [1904.04912], have been found to be an effective approach to momentum and mean-reversion trading. However, some of the key challenges in recent years involve learning long-term dependencies, degradation of performance when considering returns net of transaction costs and adapting to new market regimes, notably during the SARS-CoV-2 crisis. Attention mechanisms, or Transformer-based architectures, are a solution to such challenges because they allow the network to focus on significant time steps in the past and longer-term patterns. We introduce the Momentum Transformer, an attention-based architecture which outperforms the benchmarks, and is inherently interpretable, providing us with greater insights into our deep learning trading strategy. Our model is an extension to the LSTM-based DMN, which directly outputs position sizing by optimising the network on a risk-adjusted performance metric, such as Sharpe ratio. We find an attention-LSTM hybrid Decoder-Only Temporal Fusion Transformer (TFT) style architecture is the best performing model. In terms of interpretability, we observe remarkable structure in the attention patterns, with significant peaks of importance at momentum turning points. The time series is thus segmented into regimes and the model tends to focus on previous time-steps in alike regimes. We find changepoint detection (CPD) [2105.13727], another technique for responding to regime change, can complement multi-headed attention, especially when we run CPD at multiple timescales. Through the addition of an interpretable variable selection network, we observe how CPD helps our model to move away from trading predominantly on daily returns data. We note that the model can intelligently switch between, and blend, classical strategies - basing its decision on patterns in the data. |
| Date: | 2021–12 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2112.08534&r= |