Technical paper/S&P 500
Investment decisions driven by fine-tuned large language models and uniform manifold approximation and projection-supported clustering and hierarchical density-based spatial clustering
The author proposes an investment strategy using LLMs and text from social media posts and business and economic news and demonstrate that the strategy outperforms the chosen benchmark.
Option pricing using high-frequency futures prices
The authors examine two potential routes to improve the outcome of option pricing: extracting the variance from futures prices instead of the underlying asset prices, and calculating the variance in different frequencies with intraday data instead of…
What’s so special about time series momentum?
We find that the buy-and-hold (B&H) strategy for the S&P 500 index (^GSPC) for January 1950–April 2019 had a significantly higher return than that produced by time series momentum (TSM). However, TSM was superior in terms of the Sharpe ratio due to its…
Can shorting leveraged exchange-traded fund pairs be a profitable trade?
In this paper, the authors examine if investors can profit from the underperformance of leveraged exchange-traded funds (ETFs) in long holding periods.
Rating migrations of US financial institutions: are different outcomes equivalent?
This study employs a competing risks approach to examine the rating migrations of US financial institutions (FIs) during the period 1984–2006.
Tail-risk mitigation with managed volatility strategies
This paper examines strategy performance from an investment practitioner perspective. Using long-term data from the Standard & Poor’s 500, the authors show that these strategies offer an improvement in risk-adjusted return compared with a buy-and-hold…
Beta hedging: performance measures, momentum weighting and rebalancing effects
In this paper, the authors discuss the various performance measures of beta hedging and offer a new synthetic criterion that accounts for both risk-adjusted returns and losses of trading strategy.
Dilated convolutional neural networks for time series forecasting
In this paper, the authors present a method for conditional time series forecasting based on an adaptation of the recent deep convolutional WaveNet architecture.
Covering the world: global evidence on covered calls
Typical covered call strategies may be decomposed, using a risk and performance attribution methodology, into three components: equity exposure, short volatility exposure and equity timing. This paper applies that attribution methodology to covered calls…
News-sentiment networks as a company risk indicator
This paper defines an algorithm for measuring sentiment-based network risk, to understand the relationship between news sentiment and company stock price movements, and to better understand connectivity among companies.
Cutting Edge introduction: Hedging dependence
Hedging dependence