Skip to content
View JoshGutierrez56's full-sized avatar

Block or report JoshGutierrez56

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
JoshGutierrez56/README.md

Josh Gutierrez

MS Finance / MBA Candidate — Northeastern University D'Amore-McKim School of Business Head of Equity Research, 360 Huntington Fund ($2.1M AUM) | President, Options Club & Commodities Club | Boston, MA


Featured Research and Products

Pure News Intelligence

Evidence-linked analyst workflow for identifying disclosure changes, checking prior context, preserving institutional memory, and forming testable research questions without automatic trade recommendations.

Live Demo · Repository · Research Ideas · Universe Coverage Preview

Bounded evidence: 149 fully validated issuers and 995 evidence-backed disclosure changes; 6,190 issuers indexed for staged coverage. Two of eight Research Idea cases produced publishable hypotheses, six failed safely, and none produced an actionable trade view. Formal human validation remains pending.

Pure News Research

Reproducible empirical research testing whether filing and news features add incremental return-predictive information, including null results, preregistered gates, and public-safe synthetic fixtures.

Public Research Repository · Research Findings · Methodology

The negative empirical findings motivated the pivot from automated alpha discovery toward an evidence-linked analyst workflow.

Quantitative Research

Project Description Stack
optlab-research Factor backtesting platform — 9 signals, Russell 3000, FF6 attribution, Brinson decomposition Python, DuckDB, Polars, CVXPY
deep-momentum-network LSTM with attention mechanism, Sharpe-ratio loss, 1.65 walk-forward Sharpe PyTorch
macro-regime-hmm Hidden Markov Model built from scratch, 96.1% NBER recession detection accuracy NumPy, scikit-learn
360hf-bl-optimizer Black-Litterman + Ledoit-Wolf shrinkage optimizer for student fund CVXPY, pandas
risk-parity Risk parity portfolio optimizer with Ledoit-Wolf shrinkage and equal risk contribution CVXPY, numpy

Factor Replication Results

Factor Universe Period Sharpe L/S Source
Quality / Gross Profitability Russell 1000 2019–2023 0.443 Novy-Marx (2013)
Momentum (12-2) Russell 3000 2019–2023 0.262 Jegadeesh-Titman (1993)
Momentum (12-2) Russell 1000 2010–2024 0.139 gross Carhart (1997)
BAB / Low Volatility Russell 1000 2019–2023 -0.371 Frazzini-Pedersen (2014)
Value / Book-to-Market Russell 3000 2010–2024 -0.266 Fama-French (1992)

GP–Value cross-sectional correlation: −0.257 (p < 0.0001, n=556,200) — quality complements value.

Options Club — Fall 2026

Three-session trading game curriculum modeled on the Citi market-making card game. Each session ships with a companion GitHub repo members push to their own profiles.

Session Date Game Repo
1 — Vanilla Market Making Sept 16 Quote bid-ask on card sum mm-simulator
2 — Variance Trading Sept 30 Quote bid-ask on card variance variance-dispersion
3 — Options Payoff Oct 14 Quote bid-ask on max(sum − K, 0) greeks-payoffs

Background

7 years institutional asset management — AEW Capital Management, Loomis Sayles, Wellington Management, Prime Buchholz. Hamilton College BA.

Pinned Loading

  1. Macro-Regime-HMM Macro-Regime-HMM Public

    Gaussian HMM trained on yield curve, credit spreads, and VIX to identify Expansion/Slowdown/Crisis regimes. Baum-Welch EM + Viterbi from scratch. Shows momentum crashes in Crisis, quality is defen…

    Python

  2. Optlab-Research Optlab-Research Public

    Registry-driven equity research workbench — PIT-correct factor signals, named universes, and backtesting on CRSP/Compustat via DuckDB + Polars.

    Jupyter Notebook

  3. Cap-Rate-Decomposition Cap-Rate-Decomposition Public

    Decomposes NCREIF cap rates into risk-free rate, risk premium, and NOI growth components. Includes three-model NOI growth forecasting (OLS, ECM, ensemble), equilibrium valuation, and Bull/Base/Bear…

    Python

  4. PERE-Alternative-Data-Signals PERE-Alternative-Data-Signals Public

    Tests whether public REIT prices, Zillow rental data, and sector-specific FRED signals Granger-cause NCREIF NPI returns. Replicates AEW's finding that REITs lead private RE by ~2 quarters. Includes…

    Python

  5. REIT-PERE-Allocation-Optimizer REIT-PERE-Allocation-Optimizer Public

    Constrained mean-variance optimizer for REIT / private real estate allocation with Geltner unsmoothing, Ledoit-Wolf shrinkage, CVXPY solver, walk-forward backtest, and macro regime analysis.

    Python

  6. Relative-Value-Index Relative-Value-Index Public

    Open-data replication of the AEW Relative Value Index: Gordon Growth Model sector scoring across NCREIF NPI cap rates, NOI growth momentum, and FRED macro data. Includes walk-forward IC backtest.

    Python