ENJINE

[Paper Review] Kinetic Component Analysis - How Kalman filters work, and why they're suited to extract signals from financial datasets - de Prado and Rebonato's enhancement of the popular form of the model - Explanation of competing methods enjine.com/blog/paper-review…

What Is Machine Learning? - The two sides of ML - ML vs "traditional statistics" - Categories of ML models enjine.com/blog/what-is-mach…

[Paper Review] Deep neural networks, gradient-boosted trees, random forests: Statistical arbitrage on the S&P 500 - Explanation of deep neural networks, random forest, gradient boosted trees in layman's terms - Critique of the authors' model set up enjine.com/blog/paper-review…

Look-Ahead Bias, and Why Backtests Overpromise - How look-ahead bias is like time travelling - How the bias inflates backtests - What we can do to mitigate its influence enjine.com/blog/look-ahead-b…

Training Neural Networks: Why, As With Humans, Teaching Methods Matter - What optimization algorithms do - How Gradient Descent and successor techniques work - How the choice of algorithm impacts model performances enjine.com/blog/training-neu…

Designing Neural Networks - Origins of NN - Composition of modern NN - What it takes to create great NNs enjine.com/blog/designing-ne…

Paper Review] Algorithmic Financial Trading with Deep Convolutional Neural Networks: Time Series to Image Conversion Approach - Non-math explanation of convolutional neural networks - My critique of the paper's model - Model results put into perspective enjine.com/blog/paper-review…

Statistical Distributions and the Costliness of Hidden Assumptions - Explanation of 'fat tails' in statistical distributions - The pervasiveness of the Gaussian distribution assumption - How fat tailed reality and Gaussian assumptions clash enjine.com/blog/statistical-…

[Paper Review] Deep learning with long short-term memory networks for financial market predictions - How LSTMs work, in plain language - Reasons for the extremely high performance - Real world implementation caveats enjine.com/blog/paper-review…

[New blog post] Linear Regression: Useful, Ubiquitous, Limited enjine.com/blog/linear-regre…

[New blog post] Choi KPCA: A Tool To Help You Create Many Factor Quant Models enjine.com/blog/choi-kpca-to…

[New blog post] Machine Learning vs Quantitative Finance: What's the Difference? enjine.com/blog/machine-lear…

(New blog post) Is It Possible To Replicate Renaissance Technologies' Success? enjine.com/blog/it-possible-…

Is Low Signal to Noise Ratio Really A Problem For Financial Machine Learning? enjine.com/blog/low-signal-n…

Which portfolio optimization algorithm has been most resistant to Covid-19? We compared three algorithms to find out: enjine.com/blog/portfolio-op… #portfoliooptimization #mcos #hrp #DataScience #MachineLearning #Assetmanagement

Hear our own Peter White detail an innovative Machine Learning powered algorithm called Hierarchical Risk Parity, that could help you safeguard your portfolio enjine.com/blog/hierarchical… #portfoliooptimization #hierarchicalriskparity #mcos #MachineLearning

Check out this transcript from Steven Young's Interview with Denis Vorotyntsev, Winner of the AutoML on Time Series Regression AutoSeries Challenge enjine.com/blog/denis-voroty… @Steven_M_Young @tEarth4 #MachineLearning #DataScience #automl