“Not all those who wander are lost.”
J.R.R. Tolkien
We believe re-evaluating issues and studying nuances leads to innovative ideas, fresh perspectives and interesting discoveries. Insights encapsulate our thought processes and interim findings on the quest to answer the question “tell me something I don’t know”.
16
Feb
Robust Rolling Regime Detection (R2-RD): A Data-Driven Perspective of Financial Markets
The nonstationary and high-dimensional nature of financial markets poses significant challenges for navigation. Temporally stable regime classification offers a perspective to manage these challenges. We propose the Robust Rolling Regime Detection (R2-RD) framework that adaptively retrains with streaming data and employs temporal ensemble, label assignment, and threshold policies to address temporal instability resulting from nonstationarity, model mismatches, etc.
Since a learning-based model is only as powerful as the data it trains on, the more stable results of the R2-RD make it a better candidate for usage across AI-based applications.
16
Feb
Robust Rolling K-Means (R2K-Means): an Updateable Nonlinear K-Means Clustering Methodology for Financial Time Series
K-Means is a popular clustering algorithm designed to group data points into k clusters. In the financial industry, grouping funds or assets can isolate behaviors and define investment universes using any number of performance measures, holdings, or alternative features. Standard K-Means clustering at each time increment creates extremely unstable results due to the effects of random initialization and cluster mislabeling. Robust Rolling K-Means (R2K-Means) is the extension of K-Means to time series allowing investors to dynamically track and group funds in a stable and updateable framework.
Since a learning-based model is only as powerful as...
15
Sep
Mutual Funds – Historic Analysis: US Small Cap
Small cap mirage - running 16,500 investment strategy combinations over 20 years highlights the delusion of leveraging basic historical analysis for selecting small cap funds that consistently beat the index! Note that 20% of the funds reclassified themselves (some more than 4 times), there is marginal prediction power in performance measures (less than 1 in 5 chances), performance has struggled to beat the Russell 2000, Covid literally flipped the signals and yet, over the last decade the US Small Cap mutual fund market grew 2.5x!
For the US Small Cap mutual fund market, assuming the Russell...
01
May
US Mid Caps – Historic Analysis: Are you always wrong?
$816 billion invested and you are lucky if you get what you think you are buying! 43% of the funds have reclassified themselves (some more than 4 times), there is marginal prediction power in performance measures (less than 1 in 5 chances), performance has struggled to beat the Russell Mid Cap TR USD and yet, over the last decade the US Mid Cap mutual fund market grew 2.5x!
Overall, for the US Mid Cap mutual fund market, assuming the Russell Mid Cap TR USD as the index, there may be lesser value in looking at historical price...
25
Mar
Robust Rolling PCA (R2-PCA): Managing Times Series and Multiple Dimensions
Principal Component Analysis (PCA) is an important methodology to reduce and extract meaningful signals from large data-sets. Financial markets introduce time and non stationarity aspects, where applying standard PCA methods may not give stable results. Our robust rolling PCA (R2-PCA) accommodates the additional aspects and mitigates commonly found obstacles including eigenvector sign flipping, and managing multiple dimensions of the data-set. Since a learning-based model is only as powerful as the data it trains on, the more stable results of the R2-PCA (versus the Standard PCA) make it a better candidate for usage across AI-based applications.
17
Mar
ChatGPT Mutual Fund suggestions — Good or Bad?
Not good - buyer beware as the results were inconsistent textually and from a performance perspective, where it suggested index trackers (without specifying Indices) and otherwise generally poor performing funds. This is not surprising as it is akin to asking an English major to solve a differential equation - pun intended! The math here is in ingesting existing published results and making them contextually available and not on training the models to accurately select the asset. As such, the basis of ChatGPT is the Large Language Models (LLM) that are trained on existing ‘outcomes’ that are solicited from...
10
Mar
Portfolios – How much RoT is there?
You are pitched multiple ‘tailored’ portfolios that have seemingly very similar risk-return expectations. Beyond the regulatorily mandated disclosures, the distributors/allocators generally point to the historical performance of the portfolios and forecasted performance under scenarios. In Table 1 we take an example of five portfolios with near identical first order risk-return profiles - how would you select?
Table 1 - Model Portfolios
The baseline financial analysis of the portfolios relies on two baseline facets: (a) the performance measures being considered, and (b) the assumptions embedded in the simulations. Where,
01
Mar
US Large Cap Mutual Funds – Basic Historic Analysis: are you always wrong?
Pretty much - more than 75% of the time. Unfortunately, with reference to the SP500 TR USD you will only be selecting top performers maybe 25% of the time (depending on the measures and evaluation criteria). Irrespective, over the last decade the US Large Cap mutual fund market grew 4x despite average relative poor performance versus the Index - $8 Trillion chasing fool's gold? For the overall US Large Cap mutual fund market, assuming the SP500 TR USD as the index, there may be lesser value in looking at historical price data (and the derived performance measures) as selection criteria. As...