This is a summary of links featured on Quantocracy on Wednesday, 07/10/2019. To see our most recent links, visit the Quant Mashup. Read on readers!
-
Practical Pairs Trading [Robot Wealth]Some price series are mean reverting some of the time, but it is also possible to create portfolios which are specifically constructed to have mean-reverting properties. Series that can be combined to create stationary portfolios are called cointegrating, and there are a bunch of statistical tests for this property. Well return to these shortly. While you can, in theory, create mean reverting
-
Market Sell-off Analysis: Baseline Historical Facts [Alpha Architect]We often hear that the market is 5% off its highs or that it is down 5% from the high of the year. This alone does not tell us much. The questions I want to answer are as follows: How often does that 5% loss become a 10% loss? Or worse yet a 20% loss? In other words, what are the historical distribution of outcomes, given a loss of x%? I address this question in US markets and then
-
Day of Month and Market Timing [Alvarez Quant Trading]In my previous post, Market Timing with a Canary, Gold, Copper, LQD, IEF and much more, I tested several market timing methods. The signal was checked on the last day of the month. Now the question is what happens if we check on a different day? How different will the results be? The Test The backtest is from 1/1/2004 to 12/31/2018 on the SPY, dividends included. Buy Rule On N days before the end
-
Can You Minimize Regret By Analyzing Return Distributions? [Capital Spectator]In the grand scheme of investing, behavioral risk is second to none on the list of pitfalls that threaten to derail the best-laid plans for investing. The challenge is especially acute in the thankless task of trying to anticipate how youll react when a rough patch arrives. The mystery is all the deeper if your only experience with a fund or strategy is holding it during a bull market. There