David Ardia

Traditional trading strategies—momentum-, reversal-, or factor-based—all aim to identify and leverage recurring patterns, though these can often be challenging to define. Contrastive learning, probably the brainchild of Bo Dai, bypasses the need for explicitly knowing/naming the patterns and excels with complex, multi-modal signals. In a recent exploratory study, we (w/ Kelvin Du, Rui Mao, Erik Cambria) applied this useful deep learning approach and obtained interesting results. Read more here ⬇

22

Log-Normal Stochastic Volatility Model with Quadratic Drift by Artur Sepp & Parviz Rakhmonov. Read the full articles at https://lnkd.in/gYsqmCtx. In this paper, the authors introduce the log-normal stochastic volatility (SV) model with a quadratic drift to allow arbitrage-free valuation of options on assets under money-market account and inverse martingale measures. They show that the proposed volatility process has a unique strong solution, despite non-Lipschitz quadratic drift, and we establish the corresponding Feynman–Kac partial differential equation (PDE) for computation of conditional expectations under this SV model. #bitcoin #ethereum #nonaffinemodels #momentgeneratingfunction #cryptocurrencyderivatives #quadraticvariance #IJTAF International Journal of Theoretical and Applied Finance

41

2 Comments

#Mathematics #particularinterest 👉 A paper titled ”Distance Correlation Market Graph: The Case of S&P500 Stocks” may raise your interest. 🔗 https://buff.ly/4e2pM0e #distance_correlation; #complex_networks; #market_graph; #stock_market_network; #non_linearity; #investment_science; #graph_theory #MDPIOpenAccess #ComSciMath_Mdpi

4

🎓Last session of our Bachelier Seminar on Mathematical Finance this Friday June 21 in Paris: 🕘 At 11:15 AM with Thomas Cavallazzi (CentraleSupélec) presenting 'Propagation of chaos for mean-field systems with stable Lévy noise': 🖊️ In this talk, we will deal with McKean-Vlasov Stochastic Differential Equations (SDEs) driven by alpha-stable processes, with alpha in (1,2). We make Hölder-type assumptions on the coefficients, with respect to both space and measure variables. We will study the associated semi-group, acting on functions defined on the space of probability measures, through the related backward Kolmogorov Partial Differential Equation (PDE) describing its dynamics. We will focus in particular on its regularizing properties. The study relies on differential calculus for functions defined on the space of measures, and on Itô's formula along flows of marginal distributions of jump processes defined with Poisson random integrals. We will finally use the preceding tools to prove quantitative weak propagation of chaos for the mean-field interacting particle system associated with the McKean-Vlasov SDE. 🔗 For those interested: https://lnkd.in/gy7nMdRB ✨ See you there! #BachelierSeminar #MathematicalFinance #Research #QuantitativeFinance #Math #AppliedMathematics #PhD

48

Practical Applications of The Term Structure of Machine Learning Alpha is here: https://lnkd.in/gUi5MqhK By David Blitz, Matthias Hanauer, Tobias Hoogteijling & Clint Howard In The Term Structure of Machine Learning Alpha, from the Fall 2023 issue of The Journal of Financial Data Science, David Blitz, Matthias X. Hanauer, Tobias Hoogteijling, and Clint Howard of Robeco find that using a machine learning (ML) trading strategy designed for a three‑month predictive horizon outperformed an ML strategy with a one‑month predictive horizon by 1.58% per year for the period 2004 through 2021. Key takeaways: • For the period since 2004, machine learning (ML) models for predicting one-month-ahead stock returns have produced roughly zero net returns after costs. • ML-based trading strategies still can yield significant positive net returns by training on longer prediction horizons and using efficient trading rules. • Longer-horizon strategies select slower signals and load more on traditional asset pricing factors but still unlock unique alpha. #MachineLearning #TradingStrategies #FinancialDataScience #InvestmentResearch #PredictiveAnalytics

58

Weekend post. For those who don't know: Peter Jaeckel found an analytical formula for implied normal volatility and wrote a paper about it.   https://lnkd.in/eMxsy9_Q   We use this in #OpenSourceRiskEngine in different contexts for quite some time. However, in some ways #QuantLib seems to be the better home for the little gem, so we moved it over and it is available in the latest QuantLib release 36. It replaces the Choi approximation as the default way to retrieve implied normal vols: https://lnkd.in/eBzuhkDf   It improves accuracy by some orders.

168

3 Comments

Are markups among the main drivers of inflation in US? This article examines aggregate effect of markups on inflation. AUTHORS Sylvain LeducSylvain Leduc Huiyu LiHuiyu Li Zheng LiuZheng Liu PS: figure 3 is missing…

3

🎓Bachelier Seminar on Mathematical Finance this Friday May 24 in Paris: 🕘 At 11:15 AM with Charles Bertucci (École Polytechnique) discussing 'Stochastic optimal transport and HJB equations on the set of probability measures': 🖊️ I will introduce a stochastic version of the optimal transport problem. I'll show an analysis by means of the study of the associated Hamilton-Jacobi-Bellman equation, which is set on the set of probability measures. I will give a new notion of viscosity solutions of this equation, which yields general comparison principles, in particular for cases involving terms modeling stochasticity in the optimal control problem. Using this notion, I will be able to establish results of existence and uniqueness of viscosity solutions of the Hamilton-Jacobi-Bellman equation. These results rely on controllability results for stochastic optimal transport that I will also present. 🔗 For those interested: https://lnkd.in/gy7nMdRB ✨ See you there! #BachelierSeminar #MathematicalFinance #Research #QuantitativeFinance #Math #AppliedMathematics #PhD

59

1 Comment

🚀 Syntax Data has partnered with WEBs Investments Inc. , an innovator in volatility-managed investment strategies, to launch two distinctive new indices: the Syntax Defined Volatility US Large Cap 500 Index and the Syntax Defined Volatility Triple Q’s Index. The indices adjust exposure to the SPDR S&P 500 ETF and the Invesco QQQ ETF, respectively. 💡 Check out the full press release to learn more about the indices and their unique advantages: https://lnkd.in/eiJAPdH4

15

I agree with Yann LeCun. Current #LLMs are not intelligent, they are just knowledge accumulators of superhuman proportions. However, where I differ in view point is on the degree of threat they present today. When most of the population of major geographies has the mental age of a teenager and the level of understanding of a 6th grader, vastly superior knowledge can quickly get assimilated with intelligence. Coupling that with a quasi-permissionless access to LLM tools can result in significant risks of a non-transitory nature: - Political. That is the one typically coming to mind especially in election years and given smaller-scale precedents in 2016 - Educational. LLMs are increasingly incorporated into teaching, from a young age to university level. Pupils and students consume increasingly more content online. Content that sound right may not be, and yet may be assimilated as the truth. I have recently witnessed that first hand - Behavioral. Certain social media have documented patterns (whether voluntary or not) of triggering user dependencies. LLMs can make such patterns more scalable, whether with the goal of increasing revenue, or with the goal of manipulating target user groups Yet, I don't think that controlling #AI systems is the answer, notably because I don't think you can control technologies, and even more so through an internal group at a private entity whose upside is in pushing said technology. Supervision of companies involved in the value chain seems a lot more feasible. https://lnkd.in/gRXxHMYT

10

The June 2024 CQF program starts tomorrow. Delivered online, part-time, by a faculty of industry leaders, the program teaches essential quant finance and machine learning techniques and their application to real-world scenarios. Don't miss your chance to join: https://ow.ly/KksM50SkEto #CQF #quant #finance #quantfinance #quantitativefinance #financialengineering #career #datascience #machinelearning #riskmanagement #quantstrategies #trading #technology

23

📢 How can we enhance the efficiency of asset pricing models using ML? The paper "Kernel Trick for the Cross-Section" proposes a novel approach using the kernel trick from ML. Leveraging this technique, the study extends the set of characteristics to an arbitrary—potentially infinite—dimension, enhancing the modeling of non-linearities and interactions in characteristics. The proposed kernel-based SDF (Stochastic Discount Factor) model delivers substantial improvements: 👉 Out-of-Sample Sharpe Ratio: 3.0, compared to 1.65 with no interactions. 👉 Simulated Factor Recovery: Correlation of 0.98 using kernel-based PCA, versus 0.13 with traditional PCA. 👉 Empirical Analysis: Cross-validated Sharpe ratios improve from 0.31 to 0.75 with the Fama-French-Carhart factors, and from 1.5 to above 3 with forty anomaly characteristics. Additionally, the model maintains robust performance in out-of-sample tests, with the out-of-sample Sharpe ratio doubling compared to linear methods. The kernel method's advantage persists despite the overall deterioration of anomaly returns in recent years. 👉 Monthly Predictive R-Squared for Individual Stocks: 0.8%, compared to a benchmark of 0.2%. 👉 Daily Predictive R-Squared: 0.045%, compared to a benchmark of 0.01%. The kernel-based approach significantly enhances the efficiency and predictive power of the SDF, offering a more robust framework for understanding the cross-section of expected returns. This method demonstrates the importance of incorporating non-linearities and interactions in asset pricing models, resulting in better risk-adjusted performance and predictive capabilities. ----------------------- → Join 2500+ Finance & AI enthusiasts who receive top new research ideas weekly in their email.: https://lnkd.in/dkxSDJpq ----------------------- Link to the paper: https://lnkd.in/dhzus255

20

2 Comments

🎓Bachelier Seminar on Mathematical Finance this Friday May 31 in Paris: 🕘 At 11:15 AM with Emma Hubert (Princeton University) presenting 'A new approach to principal-agent problems with volatility control': 🖊️ The recent work by Cvitanic, Possamaï, and Touzi (2018) [1] presents a general approach for continuous-time principal-agent problems, through dynamic programming and so-called second order backward stochastic differential equations (2BSDEs). In this paper, we provide an alternative formulation of the principal-agent problem, which can be solved using more straightforward techniques, simply relying on the theory of BSDEs. 💡This reformulation is strongly inspired by an important remark in [1], namely that if the principal observes the output process X in continuous-time, she can compute its quadratic variation pathwise. While in [1] this information is used in the contract, our reformulation consist in assuming that the principal could directly control this process, in a ‘first-best’ fashion. The resolution approach for this alternative problem actually follows the line of the so-called ‘Sannikov’s trick’ in the literature on continuous-time principal-agent problems, as originally introduced by Sannikov (2008) [2]. ✅ We then show that the solution to this ‘first-best’ formulation is identical to the solution of the original problem. More precisely, using the contract’s form introduced in [1] as penalisation contracts, we highlight that this ‘first-best’ scenario can be achieved even if the principal cannot directly control the quadratic variation. 🔗 For those interested: https://lnkd.in/gy7nMdRB ✨ See you there! #BachelierSeminar #MathematicalFinance #Research #QuantitativeFinance #Math #AppliedMathematics #PhD

49