SCM

SCM Repository

[blotter] Diff of /pkg/quantstrat/sandbox/backtest_musings/strat_dev_process.Rmd
ViewVC logotype

Diff of /pkg/quantstrat/sandbox/backtest_musings/strat_dev_process.Rmd

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 1749, Mon May 1 13:34:02 2017 UTC revision 1750, Mon Jun 26 14:29:28 2017 UTC
# Line 1  Line 1 
1  ---  ---
2  title: Developing & Backtesting Systematic Trading Strategies  title: Developing & Backtesting Systematic Trading Strategies
3  author: Brian G. Peterson  author: Brian G. Peterson
4    date: "updated `r format(Sys.time(), '%d %B %Y')`"
5  bibliography: stat_process.bib  bibliography: stat_process.bib
6  output:  output:
7    tufte::tufte_handout    tufte::tufte_handout
# Line 12  Line 13 
13  keywords: quantitative trading, backtest, quantitative strategy, scientific method  keywords: quantitative trading, backtest, quantitative strategy, scientific method
14  subject: quantitative trading, backtest, quantitative strategy, scientific method  subject: quantitative trading, backtest, quantitative strategy, scientific method
15    
16  footer: Copyright 2014-2016 Brian G. Peterson CC-BY-NC-SA.  footer: Copyright 2014-2017 Brian G. Peterson CC-BY-NC-SA.
17  copyright: Copyright 2014-2016 Brian G. Peterson CC-BY-NC-SA.  copyright: Copyright 2014-2017 Brian G. Peterson CC-BY-NC-SA.
18    
19  abstract: Analysts and portfolio managers face many challenges in developing new systematic trading systems.  This paper provides a detailed, repeatable process to aid in evaluating new ideas, developing those ideas into testable hypotheses, measuring results in comparable ways, and avoiding and measuring the ever-present risks of over-fitting. ^[ *Back-testing. I hate it –- it's just optimizing over history. You never see a bad back-test. Ever. In any strategy.* - Josh Diedesch[-@Diedesch2014] ]  abstract: Analysts and portfolio managers face many challenges in developing new systematic trading systems.  This paper provides a detailed, repeatable process to aid in evaluating new ideas, developing those ideas into testable hypotheses, measuring results in comparable ways, and avoiding and measuring the ever-present risks of over-fitting. ^[ *Back-testing. I hate it –- it's just optimizing over history. You never see a bad back-test. Ever. In any strategy.* - Josh Diedesch[-@Diedesch2014] ]
20    
# Line 1706  Line 1707 
1707  - modifying existing expectations  - modifying existing expectations
1708  - track the number of trials  - track the number of trials
1709  - how do you define a 'trial'?  - how do you define a 'trial'?
1710  - CCSV sampling  - CSCV sampling
1711    - combinatorially symmetric cross validation    - combinatorially symmetric cross validation
1712  "generate $S/2$ testing sets of size $T/2$ by recombining the $S$ slices of the  "generate $S/2$ testing sets of size $T/2$ by recombining the $S$ slices of the
1713  overall sample of size $T$.  overall sample of size $T$.

Legend:
Removed from v.1749  
changed lines
  Added in v.1750

root@r-forge.r-project.org
ViewVC Help
Powered by ViewVC 1.0.0  
Thanks to:
Vienna University of Economics and Business University of Wisconsin - Madison Powered By FusionForge