This is a summary of links featured on Quantocracy on Monday, 09/19/2022. To see our most recent links, visit the Quant Mashup. Read on readers!
-
Forecasting with Decision Trees and Random Forests [Sarem Seitz]Today, Deep Learning dominates many areas of modern machine learning. On the other hand, Decision Tree based models still shine particularly for tabular data. If you look up the winning solutions of respective Kaggle challenges, chances are high that a tree model is among them. A key advantage of tree approaches is that they typically don't require too much fine-tuning for reasonable results.
-
Hierarchical PCA x Hierarchical clustering on crypto perpetual futures [Gautier Marti]PCA is a useful tool for quant trading (stat arb) but in its naive implementation suffers from several forms of instabilities which yield to unnecessary turnover (trading cost) and spurious trades. In order to regularize the model, several techniques are available: Sparse PCA Robust PCA Kernel PCA Probabilistic PCA Bayesian PCA In this blog, we will discuss one in particular: The
-
The Linear Regression-Adjusted Exponential Moving Average [Financial Hacker]There are already uncounted variants of moving averages. Vitali Apirine invented another one in his article in the Stocks&Commodities September issue. The LREMA is an EMA with a variable period derived from the distance of the current price and a linear regression line. This ensures an optimal EMA period at any point at least in theory. Will this complex EMA variant beat the standard EMA
-
Crypto PCA First Eigenvector [Gautier Marti]This short blog to illustrate an interesting fact that I found in An Analysis of Eigenvectors of a Stock Market Cross-Correlation Matrix by Nguyen and co-authors: The first eigenvector is not THE market portfolio (market-cap or uniformly weighted) as people usually believe, but a correlation-weighted market portfolio. import numpy as np import pandas as pd from scipy.stats import rankdata import
-
How Did Momentum Investing Perform After the Previous Two Valuation Peaks? [Alpha Architect]Near the end of 2021, I wrote an article noting that value portfolios looked historically cheap based on valuation spreads. I found that in the next five years (after the peak), Value investing performed quite well.(1) Following this post, I have received numerous questions related to the following question: How did Momentum investing perform after the previous two valuation peaks? This article