Skip to main content
WeiMindWeiMind
ChineseView Course
Quant Radar
23 JunTue
  1. Rob Carver：This Blog is Systematic74

    How to Automatically Identify Structural Breakpoints in Momentum Strategy Parameter Estimation

    This article introduces a method for automatically identifying structural breakpoints in the Sharpe ratio of trading rules, using independent t-tests to compare mean differences across subsamples, recursively splitting the sample while requiring each subsample to retain at least 5 years of data. The authors provide complete Python implementation code, using the momentum4 rule for CORN futures as an example, with the algorithm automatically identifying August 26, 1982 as the breakpoint.

    Read original
22 JunMon
19 JunFri
  1. Aswath Damodaran：Musings on Markets55

    SpaceX, OpenAI, and Anthropic: S&P 500 Inclusion Controversy and Investment Implications

    NYU Professor Aswath Damodaran analyzes whether SpaceX, OpenAI, and Anthropic should be included in the S&P 500 after their IPOs. He breaks down index construction, including constituent selection, weighting methods, and index unit conversion mechanisms. He notes that the S&P 500 uses free-float market capitalization weighting, and that the June 15, 2026 closing level of 7554.29 corresponds to an index unit of 0.1190.

    Read original
18 JunThu
  1. Rob Carver：This Blog is Systematic58

    Does Portfolio Optimization Need Clustering? A Backtest Examination Across Different Asset Sizes and Sample Lengths

    This is the sixth article in a portfolio optimization series. The author conducts hundreds of random backtests to examine the impact of different numbers of clusters, correlation shrinkage, SR shrinkage, and in-sample vs. out-of-sample lengths on out-of-sample Sharpe ratios across asset sizes of 20, 30, and 40. The results show that most parameter combinations have no significant differences, with only weak evidence that moderate clustering outperforms no clustering, and that the number of clusters should not be too large. The author ultimately recommends using an empirical value of 6 clusters when clear statistical evidence is lacking, while noting that clustering significantly increases optimization time.

    Read original
  2. Rob Carver：This Blog is Systematic60

    Equal-Weighted Averaging of Portfolio Weights Generated by Different Shrinkage Methods Fails to Outperform Single Shrinkage Approaches Across Multiple In-Sample and Out-of-Sample Backtests

    The author examines the effect of equal-weighted averaging of portfolio weights generated by different shrinkage methods in the Portfolio Optimization Series, comparing 9 shrinkage parameter combinations with 6 weight averaging combinations, totaling 15 approaches. The backtests use randomly selected 9 trading rules and random instruments, with 1-, 5-, and 10-year in-sample periods and 1- and 5-year out-of-sample periods, each group containing 4,000 random portfolios.

    Read original
17 JunWed
  1. Hummingbot：GitHub Releases30

    Hummingbot v2.15.0 Released

    Hummingbot v2.15.0 has been officially released. The default rate oracle source has been switched from binance to gate_io, a new Lighter connector has been added, and the functionality and stability of connectors for multiple exchanges including Hyperliquid, BingX, and Binance have been optimized. This version update includes 28 feature improvements and bug fixes, and also welcomes 3 first-time contributors.

    Read original
15 JunMon
  1. Rob Carver：This Blog is Systematic62

    Comparing Portfolio Optimization Shrinkage Methods Using Real Return Data: Backtesting Conclusions for Monte Carlo, Bootstrap, and Double Shrinkage

    In the fourth installment of the Summer 2026 Portfolio Optimization Series, the author backtests Monte Carlo, bootstrap, double shrinkage, naive mean-variance, equal weight, maximum diversification, and EPO using real trading rule and instrument return data. Each test randomly selects 9 trading rules and one instrument, comparing portfolios over 1-, 5-, and 10-year in-sample periods and 1- and 5-year out-of-sample periods. The results show that under real data, correlation estimates are less stable than Sharpe ratios, requiring more correlation shrinkage rather than more Sharpe ratio shrinkage.

    Read original
8 JunMon
  1. Robot Wealth45

    How Independent Traders Should Allocate Resources for a Triangulated Statistical Arbitrage Business

    This article is the sixth part of a series on statistical arbitrage for independent traders. It explains the three core modules of triangulated statistical arbitrage and their resource allocation priorities. High-quality pair screening is the core foundation: using 24 months of historical data, the top 10% of pairs are screened monthly and divided into 5 quality tiers, with the best tier achieving a Sharpe ratio of 1.6 and the worst only 0.42. The author discloses only the general methodology and does not reveal specific implementation details.

    Read original
1 JunMon
  1. Robot Wealth54

    Triangulated statistical arbitrage: how to identify genuine pricing deviations rather than mispricing

    This article is the fifth in a statistical arbitrage series for independent traders, focusing on the triangulated statistical arbitrage method. Using the energy sector spread network as an example, the author explains how to cross-validate multiple overlapping spreads to identify genuinely mispriced individual stocks, rather than treating deviations in a single spread as a direct trading signal.

    Read original
22 MayFri
  1. Robot Wealth53

    From Pairs to Portfolios, Part 4: A Triangulated Statistical Arbitrage Approach for Independent Traders

    This article introduces a triangulated statistical arbitrage approach. Using 8 spread groups constructed from 6 energy stocks as an example, the z-score signal of each spread is decomposed into opposite votes for the two individual stocks, which are then aggregated into long-short signals at the individual stock level. This allows direct trading of undervalued or overvalued stocks, replacing the traditional pairs-trading model of hedging both legs simultaneously. The approach can improve capital efficiency, reduce transaction costs, and increase signal utilization, while achieving variance control similar to pairs trading through a long-short basket portfolio. The article will further explain how to evaluate the quality of aggregated signals.

    Read original
19 MayTue
16 MaySat
14 MayThu
  1. vn.py：GitHub Releases47

    vn.py 4.4.0 Released

    vn.py 4.4.0 is officially released. This version adds notification functionality based on the WeChat iLink protocol, introduces the vnpy_questDB time-series database integration module, upgrades vnpy_esunny to the Esunny QD V9 API, and optimizes the vnpy.alpha data processing and backtesting workflow, while fixing multiple issues related to logging, data downloading, and algorithmic trade execution judgment.

    Read original
12 MayTue
29 AprWed
  1. Robot Wealth36

    Data mining and AI-driven 'quant by feel' are essentially the same in quantitative trading; the core should focus on the structural source of trading edge.

    The article points out that data mining and AI-driven 'quant by feel' are identical in essence and philosophy, both attempting to skip the core step of understanding how markets work and profit directly from the market. A genuine trading edge requires identifying the counterparty being profited from, the reason they continue to pay, and the rationale for one's own ability to participate in the trade; such structural edge cannot be identified through backtesting optimization alone.

    Read original
17 AprFri
  1. QuantLib：GitHub Releases22

    QuantLib 1.42.1 Released

    QuantLib 1.42.1 is a bug-fix release that removes a regression in the Gsr class which could cause an infinite notification loop. This release provides two installation packages: QuantLib-1.42.1.tar.gz and QuantLib-1.42.1.zip. The related changes were committed by @lballabio in #2543, and the final version number was set by @lballabio-bot in #2545.

    Read original
10 AprFri
  1. Robot Wealth53

    Trend or No Trend? That's the Wrong Question

    The author argues that mean reversion, trend, and momentum are merely labels for price movements rather than trading edges. A true edge requires first identifying the underlying structural cause and then validating it with data. The article uses three types of mean-reversion scenarios as examples: forced liquidation by leveraged funds, scheduled rebalancing of crypto leveraged tokens, and month-end stock-bond rebalancing by wealth management institutions. All three scenarios involve identifiable trading flows that are insensitive to price.

    Read original
9 AprThu
  1. Howard Marks Memos42

    Howard Marks Memo: Current Dynamics in the Private Credit Space

    In his latest memo, Howard Marks outlines the evolution of the sub-investment grade credit market from the 1970s to the present, focusing on analyzing the drivers behind the rapid growth of direct lending after the global financial crisis, as well as the related issues brought by the rapid deployment of large-scale capital. He points out that the development of direct lending is deeply tied to that of private equity, and the performance of private equity portfolio companies is the core determinant of whether direct loans will succeed in the future.

    Read original
29 MarSun
  1. Robot Wealth64

    Brave New Backtest: Why LLMs Have Three Structural Limitations in Trading Research

    Drawing on two academic papers, the author identifies three structural limitations of using LLMs in trading research: First, training data is saturated with trading common sense and lacks mechanism-driven edge thinking. Second, across 35+ models, accuracy in time-series value tracking declines logarithmically as the number of historical updates increases, and neither prompt engineering nor expanding the context window can fix this. Third, outputs from 25+ models on open-ended questions show 71%–82% similarity, and models with different architectures converge to the same common answers.

    Read original
24 MarTue
23 MarMon
  1. Robot Wealth54

    AI Will Create Millions of Quantitative Traders, Most of Whom Will Lose Money

    The author notes that AI has dramatically lowered the barrier to developing quantitative strategies, enabling fully coded and backtested strategies to be generated within minutes. However, cheap experimentation amplifies the multiple-testing problem, producing a flood of false backtest results that appear effective but are actually overfit to noise.

    Read original
20 MarFri
27 FebFri
  1. PyPortfolioOpt：GitHub Releases20

    PyPortfolioOpt v1.6.0 released

    PyPortfolioOpt v1.6.0 has been officially released, adding support for Python 3.14 and pandas 3.0, while dropping support for Python 3.8 and 3.9. This version reduces dependency size; existing dependencies can be used as optional dependencies via the newly added all_extras dependency set. It also optimizes CLA calculation speed and includes multiple maintenance and documentation updates.

    Read original
7 JanWed
  1. Bridgewater: Research & Insights55

    Bridgewater Research: The Macroeconomic Impact of the AI Capital Spending Boom

    Bridgewater released a research report on January 7, 2026, estimating that AI capital spending will contribute about 140 basis points to U.S. GDP growth in 2026 and about 150 basis points in 2027, a scale comparable to the contribution of business investment to growth during the dot-com bubble.

    Read original
24 DecWed
26 NovWed
  1. Bridgewater: Research & Insights42

    The Release of Google Gemini 3 Signals the Start of an AI Industry “Resource Grab” Phase

    Bridgewater Co-Chief Investment Officer Greg Jensen and AIA Labs Chief Scientist Jas Sekhon released a viewpoint arguing that Google Gemini 3 is currently the strongest publicly available general-purpose large AI model, with pre-training compute at least 2–3 times higher than GPT-4o, validating that the path of scaling pre-training remains sustainable. This development will drive the AI industry to continue increasing investment in compute, chips, and power, with significant implications for capital markets and the economy.

    Read original
13 NovThu
  1. Bridgewater: Research & Insights50

    Bridgewater Research: How Canada Is Responding to U.S. Mercantilism

    Bridgewater released a study on November 13, 2025, analyzing Canada's response path under U.S. high tariffs and USMCA uncertainty. The report notes that Canada's GDP contracted in Q2 2025, after which it adopted a three-pillar strategy of trade diversification, fiscal expansion, and loose monetary policy, aiming to double non-U.S. exports within a decade and add over CAD 300 billion in trade volume. The fiscal deficit for FY2025-26 reached CAD 78 billion, accounting for 2.5% of GDP. The Bank of Canada maintains an accommodative stance, and initial signs of stabilization have emerged in manufacturing orders, residential construction, and the labor market.

    Read original
4 NovTue
  1. EDGAR 13F：Scion16

    Scion (Michael Burry) filed its 13F-HR holdings report for the quarter ending September 30, 2025

    Scion (Michael Burry) filed its 13F-HR quarterly holdings report on November 3, 2025, for the reporting period ending September 30, 2025. The report consists of 2 files, with the submission number 0001649339-25-000007, and the institution's registered address is located at 20665 Fourth Avenue, Suite 201, Saratoga, California.

    Read original
16 OctThu
  1. Bridgewater: Research & Insights46

    Power Politics: Energy Self-Sufficiency in a Modern Mercantilist World

    Bridgewater released a research report outlining the energy supply and security landscape across the U.S., China, and Europe in the context of modern mercantilism. The report notes that cheap and abundant electricity is critical to the AI race and national competitiveness, and governments are now making energy self-sufficiency a core policy priority. The report also analyzes the differentiated challenges and advantages of the three regions in energy supply chains and power capacity expansion.

    Read original
22 SeptMon
  1. Bridgewater: Research & Insights52

    Bridgewater: Inflation Risks Enter a New Era of Structural Uptrend

    Bridgewater released a research report, arguing that the inflation center in the post-pandemic era has risen compared with that before the pandemic, 2% is more like a bottom than an upper limit, and the distribution of inflation outcomes is wider and skewed upward. The report compares the current environment with that in the 2010s, pointing out that restraining factors such as China's export of low-priced goods, shale oil production, and post-crisis deleveraging have weakened; while the medium-term expansion deficit of about 7% of US GDP, the compromised independence of the Federal Reserve, tariffs and trade restructuring, potential depreciation of the US dollar, and AI power and commodity demand constitute structural inflationary pressures.

    Read original
10 JulThu
  1. Bridgewater: Research & Insights34

    What Does It Take to Revive U.S. Manufacturing? Part 2: Boosting U.S. Manufacturing Productivity

    Bridgewater, in collaboration with Eclipse, released the second part of its series on reviving U.S. manufacturing, pointing out that U.S. manufacturing lacks competitiveness due to high labor costs and a weak industrial base, and tariffs cannot bridge the cost gap. The report proposes leveraging mature technologies such as automation, computer vision, and additive manufacturing, as well as cutting-edge technologies like generative AI and embodied AI, to boost manufacturing productivity by 1 to 2 orders of magnitude. Scaling the application of these technologies is expected to take nearly a decade.

    Read original
2 JulWed
  1. Bridgewater: Research & Insights42

    What Does It Take to Revive U.S. Manufacturing? Part 1: Structural Headwinds Facing the United States

    Bridgewater, in partnership with industrial technology investment firm Eclipse, released a study exploring the structural challenges of reviving U.S. manufacturing. Currently, about 40% of goods consumed in the U.S. rely on imports, with a manufacturing job gap of nearly 5 million over the past 20 years. For some product categories, even a 25% tariff cannot close the gap between domestic and foreign production costs. U.S. manufacturing productivity has stagnated for 15 years, and boosting productivity is the core path to enhancing the competitiveness of domestic manufacturing.

    Read original
19 MarWed
  1. Bridgewater: Research & Insights55

    Bridgewater Research: Is the Green Energy Transition Dead?

    Bridgewater released a research report on March 19, 2025, noting that energy policies in major global economies are shifting from a climate-first approach to prioritizing energy security and industrial competitiveness. The current energy landscape is characterized by incremental additions rather than a transition-driven replacement, with fossil fuels and renewable energy growing simultaneously, though the pace of decarbonization will slow.

    Read original
17 SeptTue
  1. Bridgewater: Research & Insights54

    Bridgewater Research: Drivers of U.S. Stock Outperformance and Conditions Needed for Future Repeats

    Bridgewater research indicates that since 2010, the excess returns of U.S. stocks relative to developed markets have been roughly equally driven by revenue growth, margin expansion, and rising P/E multiples, with a stronger U.S. dollar further boosting returns for unhedged global investors. The technology sector contributed 54% of the excess returns, of which 22% came from index overweights in technology and 32% from U.S. tech companies significantly outperforming their global peers.

    Read original
10 SeptTue
  1. Bridgewater: Research & Insights45

    Bridgewater Research: Is the AI Bubble Ahead of Us or Behind Us?

    The Bridgewater research team believes the AI bubble has not yet arrived, noting that the transformative technology and mature macro monetary environment needed to fuel a major bubble are already in place. This view addresses market concerns about AI valuations, commercialization, and slowing technological progress, while also highlighting risks of AI misuse and regulatory uncertainty. Bridgewater established AIA Labs in early 2023 with the goal of building a fully AI-driven investor that can outperform humans.

    Read original