Trading – Ugur Candan

News-Based Trading ML Experiments – Technical Setup

Posted on 28/06/202628/06/2026 by Ugur

Extracting Alpha from the News Cycle This documentation details a machine learning research project conducted in 2026 to evaluate how global news sentiment and economic calendars predict currency and commodity price movements. Researchers utilized the GDELT Global Knowledge Graph and economic event data to engineer fifteen unique features, testing them across 72 experimental configurations using gradient-boosted models like CatBoost and XGBoost. The study found that USDJPY was the most responsive instrument to news signals, with the highest-performing models achieving a Sharpe ratio of +2.653. Results indicated that a 24-hour prediction…

Analysis of ML Signals

Posted on 28/06/202628/06/2026 by Ugur

https://ugurzafercandan.github.io/UTradeDashboard Open Orders list is showing a real time portfolio outcome. Majority of the signals are performing worse than anticipated back testing.I have results for 1 months of future test trading. Only 4 is showing positive returns. Rest is having issues in maintaining positive returns. I will continue monitoring

Forex News Trading with Machine Learning

Posted on 09/05/202609/05/2026 by Ugur

1 source·May 9, 2026 This research examines how machine learning can be used to systematically trade the foreign exchange market by analyzing macroeconomic news events. High-impact reports like Nonfarm Payrolls and interest rate decisions create predictable price patterns that can be exploited through sentiment analysis and deep learning models. The text highlights specialized tools like FinBERT and LLM-powered agents that process news data to predict currency movements and optimize trade execution. Beyond technical modeling, the…

DEEP RESEARCH: Deviation from Fair Value in Commodity & FX Market Efficiency (2020-2026)

Posted on 21/04/202621/04/2026 by Ugur

EXECUTIVE SUMMARY DIRECT ANSWER: Commodity and FX markets exhibit persistent and exploitable deviations from theoretical fair value due to institutional constraints, behavioral biases, market microstructure frictions, and information processing limits. Key exploitable inefficiencies include: PPP deviations persisting 3-7 years, commodity futures term structure roll yields of 5-15% annually, BEER/FEER exchange rate misalignments of 15-40%, and oil price deviations from marginal cost by $20-40/bbl during supply shocks. CONFIDENCE: Medium-High – Based on 60+ sources including IMF/ECB…

Commodity Market Inefficiencies: Deep Research Report

Posted on 20/04/202620/04/2026 by Ugur

Cross-Asset Arbitrage, Structural Spreads, and Actionable Alpha Opportunities Prepared: April 19, 2026Analyst: Quantitative Research — Commodity Markets \& Cross-Market InefficienciesScope: Global commodity markets, derivatives, FX, and emerging asset classes — 1. Executive Summary This report identifies and analyzes 21 market inefficiencies spanning energy, metals, agriculture, FX/macro, emerging, and derivatives markets. Six original inefficiencies are extended with 2023–2026 data, and 15 newly identified opportunities are profiled in detail. Key findings: Top 5 most actionable today (2026):…

Test Your Trading vs Investing Knowhow

Posted on 21/12/202521/12/2025 by Ugur

If you do not understand what is going on below on a “Gamma scalping” strategy on BTC (bitcoin as an asset) You should never ever try trading on any asset class 🙂 How a real fund makes money from volatility (step-by-step, using $100,000 BTC) Assume: At the start:Delta-neutral. No directional risk. Now let’s see how they profit. Step 2 – BTC goes up 10% → $110,000 Straddle delta becomes +0.5 BTC.The fund is unintentionally long…

5 Year Analysis of US30, GER30, NIKKEI, CHINA50, WTI, UKX, US500, EmergingBasket, GOLD, Bitcoin, Meta, SAP, Nvidia, Palantir, Google

Posted on 29/11/202507/02/2026 by Ugur

Lets see them in Order: 5 Year Analysis of US30, GER30, NIKKEI, CHINA50, WTI, UKX, US500, EmergingBasket 5 Year Analysis of US30, GER30, NIKKEI, CHINA50, WTI, UKX, US500, EmergingBasket ADD GOLD. 5 Year Analysis of US30, GER30, NIKKEI, CHINA50, WTI, UKX, US500, EmergingBasket, GOLD ADD Bitcoin. 5 Year Analysis of US30, GER30, NIKKEI, CHINA50, WTI, UKX, US500, EmergingBasket, GOLD, Bitcoin ADD (Meta, SAP, Google) as an example of top europe and US stock tickers. Now…

USA Stocks Trading Algo Results: Golden Question: Which Yield Curve is better?

Posted on 22/11/202522/11/2025 by Ugur

Let’s see your peference. I have posted 4 Different Algo’s. they more or less do the same but the yield curves are very different. I have also linked the code blocks if you want to test it yourself.

What is the CAGR achieved in Trading by TOP Trader “James Simons”?

Posted on 15/07/202515/07/2025 by Ugur

Yes James Simons… I have been referring to him on different occasions. This math lover, who earned his PhD from the University of California, is known for founding and creating Renaissance Technologies Chern-Simons formwith a technical analysis style in its trading activities. Reported revenue reached $18.5 billion in 2018. Although James Simons was included in Bloomberg’s 2011 list of the 50 most influential people and was named “the world’s smartest billionaire” by the Financial Times…

Long Waited 156 Signal Analysis MQL5 03/2025

Posted on 01/04/202515/07/2025 by Ugur

Lets Assume we bet 100 USD on Each signal. and I would stop trading the signal at 25% DD.156 x 100 = 15,600 Total First Selection of Signals: 156 (I counted them wrong as 157) 33 :Exploded 🙂 – account closed. We do not know the exact reason for stopping but we assume they blow the account up. We would have stopped at 25% DD and we have 75 USD after stoploss. 70 of them…

1 2 »