Sports Analytics Weekly by kubeia.io - 32/2023
- Week 52
- Week 51
- Week 49
- Week 48
- Week 47
- Week 46
- Week 45
- Week 44
- Week 43
- Week 42
- Week 41
- Week 40
- Week 39
- Week 38
- Week 37
- Week 36
- Week 35
- Week 34
- Week 33
- Week 32
- Week 31
- Week 30
- Week 29
- Week 28
- Week 27
- Week 26
- Week 25
- Week 24
- Week 23
- Week 22
- Week 21
- Week 20
- Week 19
- Week 18
- Week 16
- Week 14
- Week 13
- Week 12
- Week 10
- Week 9
- Week 6
- Week 5
- Week 4
- Week 3
- Week 1
đź’° Quantitative Finance
The beta anomaly and the quality effect in international stock markets - ScienceDirect
sciencedirect.com - Reza Bradrania, Jose Francisco Veron, Winston Wu
We investigate the beta anomaly and its relationship with stock quality in international stock markets. The beta anomaly exists in three aggregates (Europe, Pacific, and Global) and fourteen of the twenty-two country portfolios. We further demonstrate that stock quality explains the beta anomaly in international markets. The beta anomaly is statistically significant among junk (low-quality) stocks, and it does not exist among quality (high-quality) stocks. The results are robust in portfolio and regression analyses, both before and after controls. Finally, we show that the alphas of the beta anomaly estimated using the Fama–French–Carhart factor as well as Fama–French five-factor models disappear when augmented by the quality-minus-junk (QMJ) factor.
🤖 Machine Learning
Understanding Power Analysis
twitter.com - Selçuk Korkmaz
What's Power Analysis?
Power analysis helps determine the sample size required for a study, ensuring that it can robustly detect an effect if one exists. Essentially, it's about ensuring you've got enough data to make solid conclusions.
Understanding non-Bayesian
princeton.edu - Chris Sims
Once one becomes used to thinking about inference from a Bayesian perspective, it becomes difficult to understand why many econometricians are uncomfortable with that way of thinking. But some very good econometricians are either firmly non- Bayesian or (more commonly these days) think of Bayesian approaches as a “tool” which might sometimes be appropriate, sometimes not. This paper tries to articulate the counterarguments to a Bayesian perspective. There are some counterarguments that are frequently expressed, but are not hard to dismiss. Others, though, corre- spond to types of application where convenient, seemingly sensible, frequentist tools exist, while Bayesian approaches are either not yet developed or seem quite incon- venient. And there are also counterarguments that relate to deep questions about inference on infinite-dimensional parameter spaces and to corresponding pitfalls in the application of Bayesian ideas. Section II explains the difference between Bayesian and frequentist approaches to inference. Section III discusses commonly heard, but weak, objections, while section IV takes up subtler issues. Section V illustrates the subtler issues in the context of some specific models.
Statistical Shrinkage (2)
eranraviv.com - Eran Raviv
During 2017 I blogged about Statistical Shrinkage. At the end of that post I mentioned the important role signal-to-noise ratio (SNR) plays when it comes to the need for shrinkage. This post shares some recent related empirical results published in the Journal of Machine Learning Research from the paper Randomization as Regularization. While mainly for tree-based algorithms, the intuition undoubtedly extends to other numerical recipes also.
Machine Learning on Tabular Data
manning.com - Mark Ryan and Luca Massaron
Business runs on tabular data in databases, spreadsheets, and logs. Crunch that data using deep learning, gradient boosting, and other machine learning techniques.
Portfolio Management: A Deep Distributional RL Approach
ssrn.com - David Pacheco Aznar
This thesis presents the development and implementation of a novel Deep Distributional Reinforcement Learning (DDRL) approach in the field of quantitative finance: the Distributional Soft Actor-Critic (DSAC) with an LSTM embedding. The model is built to further stabilize the performance of the widely used deep reinforcement learning model Soft Actor Critic (SAC) and is compared against traditional baselines such as Hierarchical Risk Parity, Minimum Variance Portfolio, DJIA and equal weight portfolio. The results show increased returns with less risk associated and stability over Soft Actor Critic and traditional baselines both in random path validation and backtest with daily frequency. The distributional component allows the model to incorporate an inherent sense of risk. The embedding enhances the temporal-dependency awareness and the observation space is composed of multiple features based upon past returns. Thus, this thesis opens the door to further research in the use of deep distributional reinforcement learning models in the context of finance.
🕰️ Blast From the Past
Shrinkage in statistics
eranraviv.com - Eran Raviv
Shrinkage in statistics has increased in popularity over the decades. Now statistical shrinkage is commonplace, explicitly or implicitly.
But when is it that we need to make use of shrinkage? At least partly it depends on signal-to-noise ratio.