Quant Research Report

Published on Jul 10, 2024. This study investigates how financial time series forecasting can be improved using auto- regressive and regression models with a Bayesian approach. Specifically, it focuses on forecast- ing the maximum logarithmic return for five-step-ahead data on the S&P 500 ETF, EUR/USD, and Bitcoin. The dataset spans from January 1, 2021, to June 15, 2024, with 10-minute and 30-minute intervals. The findings suggest that the models outperform the ARIMA-GARCH benchmark when applied to S&P 500 ETF and Bitcoin, which are characterized by their fat- tailed distribution and extreme outliers. Moreover, the results show superior performance when the models are employed in a long-only trading strategy, especially during choppy and bear markets.

Published on Jun 7, 2024. Pair Trading, a popular statistical arbitrage strategy, involves shorting overvalued assets and buying undervalued ones to profit from their return to equilibrium. This research explores and compares aspects of pair trading across pair formation and trading stages. Multivariate trading pairs are built by employing the OPTICS algorithm, empirical study reveals that diversification characteristics of multivariate pairs enhance risk-adjusted returns, but they may not necessarily outperform univariate pairs in dynamic market conditions. Furthermore, a forecasting-based trading framework is proposed to capture short-term trend reversals and improve risk-adjusted returns. Among the forecasting methods employing the ARIMA, XGBoost, and LSTM models, the ARIMA models demonstrate the best forecasting and trading performance, achieving a notable 23.038% annualized return, a 5.612 Sharpe ratio, and an over 80% win rate throughout the testing period.

Published on Dec 28, 2023. Return forecasting is a widely used techniques in trading. Traders and investors alike employ various models and tools to predict future returns, aiming to gain a competitive advantage. On the other hand, volatility forecasting is a commonly utilized method within option trading since the value of option is heavily influenced by volatility, given the inherent nature of options. Thus, accurately predicting future volatility becomes crucial. However, what if we were to combine both return and volatility forecasting in option trading? Could this integration result in more refined and effective trading strategies? Let's explore this fascinating idea.

This research aims to explore the integration of combining return and volatility forecasting within option trading and examine its effects. We will employ various forecasting models in conjunction with different hybrid trading systems to assess the performance of different approaches and explore the potential opportunities they offer.

Published on Oct 12, 2023. Portfolio diversification is a challenging task that involves the careful selection of clusters and allocation of weights. The selection of clusters involves identifying time series that exhibit similar characteristics or behaviors, such as correlation patterns or risk profiles, but are traditional approaches useful in determining the clustering effectiveness and consistency?

Our research aims at identifying the optimal clustering, dimensional reduction techniques, and weighting methods that can yield increased consistency in clustering results, thus leading to more effective risk diversification. We also aim to investigate whether learning latent representation before applying these techniques could enhance our results, using the Sharpe ratio as a measure of risk-adjusted return.

Published on Oct 05, 2023. Which asset classes have historically demonstrated strong performance in disinflationary conditions? With the decline of inflation rates since mid-2022, significant shifts have occurred in the economic landscape.

It is commonly believed that commodities encounter challenges while the value of the dollar is anticipated to increase in disinflationary environments. Do these hypotheses align with historical data? How do other asset classes perform in disinflationary environments?

This research aims to investigate the asset classes and strategies that exhibit robust performance during disinflation. We will analyze the historical effects of disinflationary periods on major asset categories such as stocks, bonds, commodities, and currencies and see if they align with our intuitive expectations.

Published on Sep 28, 2023. Pair trading is a market-neutral strategy that involves constructing a long-short portfolio comprising two highly correlated assets. The primary objective of this strategy is to generate profits irrespective of market movements.

While the long-short portfolio effectively mitigates most market risks, there are still certain risks that can be attributed to the factor model. Therefore, it is worth considering whether the utilization of a factor model during the portfolio construction process can enhance market neutrality even further.

Published on Jul 3, 2023. This study seeks to leverage the power of Python programming and statistical methodologies to conduct an analysis and prediction of financial market volatility and price movement. The goal is to unravel the complexities of volatility, exploring its inherent risks and potential opportunities for traders and investors. This research endeavors to serve as a cornerstone for individuals and organizations, assisting in refining investment strategies, risk management approaches, and financial forecasting processes.

Published on May 14, 2023. How can the process of pair selection be enhanced through the integration of machine learning techniques with statistical methods? In this study, we aim to identify pairs of stationary stocks in the S&P 500 universe by applying statistical tests and clustering methods with machine learning techniques. Our research compares various algorithmic techniques such as NSGA-II, Soft DTW, GMM, and DBSCAN for clustering, with filtering of stationary stocks using ADF Value and Hurst Exponent. Furthermore, we examine whether utilizing these filtering techniques leads to an improved trading performance when employing the Kalman spread strategy and explore the correlation between the obtained results and stationary metrics.

Published on Apr 26, 2023. If we can equalise risk contribution by asset class, why can't we do the same for market regimes? In this publication, we aim to provide insights into the All-Weather Portfolio (ALP) and test whether we can use this insight to construct a resilient portfolio that equalizes its risk over different market environments. We would first provide insights into the rationale behind the standard ALP and asset class performance under different market regimes, then we would attempt to reconstruct the ALP along with the core concept proposed.

Published on Apr 8, 2023. Can investors predict G10 spot price by inflation rate and interest rate data? After studying International Parity Condition on G10, we would like to further study the market directional insights based on inflation rate and interest rate data. We examine the performance of the trading strategy by incorporating XGBoost from Machine Learning as the prediction model and generating the trading signal from it. Last by not least, we would study the importance of inflation rate and interest rate and PCA enhancement schema on XGBoost.

Published on Mar 23, 2023. Should investors follow popular trading strategies? This research report aims to explore whether trading strategies based on three well-known market anomalies can still yield profitable results in recent years. By analyzing the performance of trading strategies based on these anomalies in recent years, we will assess whether they can provide investors with a means of generating alpha.

Published on Mar 8, 2023. International Parity Condition is an economic theory that suggesting the dependency of the future foreign exchange spot rate on price level and interest rate. However, the assumptions behind the theories make the expected spot rate unrealistic. In this report, we would like to examine the deviations in the real world G10 currency pairs and see whether we can extract some insights from the theoretical value based on the Parity Conditions.

Published on Feb 20, 2023. Options have often been described as financial tools to help reduce risk. Yet, the use of such instruments could have detrimental effects on the risk-adjusted returns of a portfolio. We will investigate the practice of purchasing or writing options instead of implementing a buy-and-hold strategy. We will show that such acts would reduce Sharpe Ratios, assuming that stock prices are random.