R

An updated version of xts is now on CRAN. The most notable change is that plot.xts() now supports a log scale y-axis. This involved a significant refactor of the plot.xts() internals, so it’s possible to have introduced some bugs.

This post extends the replication from the Adaptive Asset Allocation Replication post by running the analysis on OOS (out-of-sample) data from 2015 through 2023. Thanks to Dale Rosenthal for helpful comments.

The paper, “Adaptive Asset Allocation: A Primer” by Adam Butler, Mike Philbrick, Rodrigo Gordillo, and David Varadi addresses flaws in the traditional application of Modern Portfolio Theory related to Strategic Asset Allocation. It shows that estimating return and (co)variance parameters over shorter time horizons are superior to estimates over long-term horizons because parameter estimates vary substantially over time. Longer-term estimates do not account for this variability in the short-term. They propose an Adaptive Asset Allocation portfolio construction methodology that uses the new parameter estimates to substantially improve performance relative to Strategic Asset Allocation.

An updated version of quantmod is now on CRAN. It includes an awesome new feature that allows you to import up to 7 days of intraday data from Yahoo Finance!

quantmod and getSymbols() have been a core part of the R/Finance ecosystem for over 15 years. We want to change some things, but they would break existing code. We can make these changes in the new ‘rfimport’ package instead.

An updated version of xts is now on CRAN. This release patches a few issues with the features added in version 0.13.0 and addresses a few maintenance issues that popped up recently.

An updated version of quantmod is now on CRAN. It adds functions HL(), is.HL(), and has.HL() to check for ‘high’ and ’low’ price columns. It also makes accessing Yahoo Finance price, dividend, and split data more robust. getSymbols.FRED() got to and from arguments, like other getSymbols() methods. The remaining changes are bug fixes and maintenace chores.

An updated version of xts is now on CRAN. This release adds several exciting changes: open-ended time-of-day subsetting, smarter conversions to xts from data.frames/data.tables/tibbles; to.period() handles custom endpoint values, print() truncates rows like data.table, and str() provides more informative output. There are also changes to make xts more consistent with zoo, some minor speed improvements, and the usual smattering of bug fixes.

An updated version of xts is now on CRAN. This release is a big one, with lots of changes. Plotting functionality got a lot of attention. Another notable change is that merge.xts() now supports suffixes. Plus the obligatory bug fixes and refinements to make xts more robust.

I follow Quantocracy on Twitter, and I found Rolling mean correlation in the tidyverse by Robot Wealth. They say to let them know if you’d approach it differently. I would, so I thought it would be interesting to replicate the analysis using tools I’m familiar with: xts and TTR.