strand: A framework for investment strategy simulation

CRAN status Build Status codecov


strand provides a framework for performing discrete (share-level) simulations of investment strategies. Simulated portfolios optimize exposure to an input signal subject to constraints such as position size and factor exposure.

The package vignette provides an in-depth discussion of setup and usage. See vignette("strand").


Tiingo and EDGAR sample data

The package’s sample data set includes value and size factors, as well as daily pricing information, for most of the stocks in the S&P 500 for the period June-August 2020. The source of fundamental data is EDGAR, and all pricing data was downloaded using the Tiingo Stock API. Special thanks to Tiingo for giving permission to release this data.


# Install the latest version from CRAN:

# Install development version from GitHub using remotes:

Note on solvers

The strand package uses GLPK as the default solver for portfolio optimization. As a result, it depends on package Rglpk.

It is possible to use SYMPHONY instead, by setting solver: symphony in the simulation’s configuration file and installing the Rsymphony package. Note that you will need to install SYMPHONY on your system first, and on OS X perform a few extra steps to install Rsymphony.


Four ingredients are required to run a simulation using strand:

  1. Configuration file. A file in yaml format that describes the parameters of the simulation, such as the input signal, risk constraints, trading limits, position limits, the location of data inputs, etc.

  2. Security reference. A listing of all securities allowed in the simulation and any categorical values (such as sector and industry) that can be used in exposure constraints.

  3. Signal, factor, and supplementary data. Data for each day including the input signal (to which exposure is maximized) and any factors that appear in constraints. Supplementary data could include, for example, a daily measure of market capitalization for use in universe construction.

  4. Pricing data. Daily prices, dividends, and trading volume for computing market values and filling orders. Unadjusted prices and accompanying adjustment ratios may be used.


# Load up sample data

# Load sample configuration file
config <- example_strategy_config()

# Create the Simulation object and run
sim <- Simulation$new(config,
                      raw_input_data = sample_inputs,
                      raw_pricing_data = sample_pricing,
                      security_reference_data = sample_secref)

# Print overall statistics
##                            Item   Gross       Net
## 1                     Total P&L -32,355   -40,867
## 2       Total Return on GMV (%)    -1.8      -2.3
## 3  Annualized Return on GMV (%)    -6.9      -8.7
## 4            Annualized Vol (%)     8.9       8.8
## 5             Annualized Sharpe   -0.77     -0.98
## 6              Max Drawdown (%)    -6.6      -6.8
## 7                       Avg GMV         1,976,540
## 8                       Avg NMV              -716
## 9                     Avg Count               189
## 10           Avg Daily Turnover            53,022
## 11      Holding Period (months)               3.6

Example shiny application (local)

To run an example shiny application that allows interactively configuring and running a simulation:


Example shiny application (docker)

If you have docker and docker-compose installed, you can run the example shiny application by cloning the github repository and running the following commands from the top-level directory:

$ docker-compose build
$ docker-compose up

The application will run by default on port 80. To configure edit docker-compose.yml.