Building an R-Based Cross-Sectional Momentum Study
Summary
The document describes a planned empirical study of cross-sectional stock momentum using OSEAX constituents over a historical sample. Its proposed strategy ranks assets by past performance over a selected horizon, buys the strongest group, and shorts the weakest group, following an approach associated with Titman. The author is new to R and asks for guidance on programming and relevant learning materials.
The responses point toward several types of resources: an R tutorial focused on momentum, books on quantitative finance and general R programming, and a text on time-series analysis using R. One reply also suggests looking for an existing backtesting script. These are recommendations rather than a worked implementation or evaluation of the strategy. The document provides no test results, detailed portfolio rules, transaction-cost treatment, or discussion of survivorship and data-quality issues, so it is best read as a starting point for research resources rather than evidence about momentum performance.
Key ideas
- The proposed study ranks stocks by past returns and takes long and short positions in the strongest and weakest groups.
- The author plans to use OSEAX data covering 1980 to 2014.
- Suggested learning materials span R programming, quantitative finance, and time-series analysis.
- The discussion offers resource pointers but no implemented backtest or empirical findings.
Tags
Full text
# Thesis using Momentum strategies in R, tips on good books, guidelines etc on how to do the programming? # Thesis using Momentum strategies in R, tips on good books, guidelines etc on how to do the programming? I am quite new to R and will be doing an empirical analysis of momentum strategies in R using a dataset from the index OSEAX from 1980 to 2014. The momentum strategy will for the most part resemble Titman (1993) where we go long on the 30% top performing assets for a given horizon and short the 30% losing stocks. So far I have found this resource (part 1-5): https://rbresearch.wordpress.com/2012/08/23/momentum-with-r-part-1/ Any good R specific finance books og other resources that might help? ## Answer by choma (score 1) https://quant.stackexchange.com/a/19549 Introduction to R for Quantitative Finance received a favorable review here: http://www.thertrader.com/category/book-review/ Besides finance-specific books, perhaps 'R Cookbook'? ## Answer by Robert Szóstakowski (score 1) https://quant.stackexchange.com/a/19552 "The Art of R Programming (A Tour of Statistical Software Design)" by Norman Matloff. It has quite high marks on Amazon. Moreover, you can find a legal version of this book on the Internet. ## Answer by Drew (score 1) https://quant.stackexchange.com/a/20976 This book by Shumway and Stoffer (two Pitt Stats profs) is excellent IMO: Time Series Analysis and Its Applications: With R Examples (Springer Texts in Statistics): 9781441978646 http://www.amazon.com/Time-Series-Analysis-Its-Applications/dp/144197864X ## Answer by Richi Wa (score 0) https://quant.stackexchange.com/a/21268 You got material about momentum, you got material about R ... usual definitions of momentum are not that difficult ... what are you looking for? A programmed back testing script? Go here and you find it all.
Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.