hraoyama/radish

Full backtesting and research system in python with fictional OMS, risk managers, etc. Layout: https://github.com/hraoyama/radish/blob/master/resources/docs/diagrams/data_structure_radish_plan.jpg

radish

Backtesting and research platform

Required

For radish framework

1. install the following:

• Anaconda python 3.6+ with pyCharm IDE and jupyter notebook
• https://arctic.readthedocs.io/en/latest/ (will require mongodb installation as well but detailed in above)
• https://redis.io/
• https://www.atlassian.com/git/tutorials/install-git
• https://confluence.atlassian.com/get-started-with-sourcetree/install-sourcetree-847359094.html
• http://kdiff3.sourceforge.net/

2. get a free github account
3. get to know some github: https://www.youtube.com/watch?v=HVsySz-h9r4

For notebooks

1. install R (rproject.org), make sure to have an R_HOME environment variable (e.g. C:\Program Files\R\R-3.4.3) and an R_PATH environment variable (e.g. %R_HOME%\bin\x64).

2. install Rtools. In R: install.packages("installr");require(installr);install.Rtools() see https://www.rdocumentation.org/packages/installr/versions/0.22.0/topics/install.Rtools

3. Edit your enviroment variables to have an R_TOOLS variable to have the Rtools path (e.g. C:\Rtools) and an R_TOOLS_PATH with the rtools subdirectories (e.g. %R_TOOLS%\bin;%R_TOOLS%\gcc-4.6.3\bin)

4. open an anaconda prompt and install the relevant packages:pip install rtools followed by conda install rpy2. rpy2 does not seem to play well with the pip installation, conda seems to work fine for some reason.

5. open a python notebook and execute:%load_ext rpy2.ipython. The default for the notebook is python, if you want to use R the new cell in the notebook should start with %%R.

```
%load_ext rpy2.ipython
import pandas as pd
df = pd.DataFrame({'cups_of_coffee': [0, 1, 2, 3, 4, 5, 6, 7, 8, 9],'productivity': [2, 5, 6, 8, 9, 8, 0, 1, 0, -1]})
```

%%R -i df -w 5 -h 5 --units in -r 200
install.packages("ggplot2",quiet=TRUE)
library(ggplot2)
ggplot(df, aes(x=cups_of_coffee, y=productivity)) + geom_line()

The second cell imports df from global environment and makes the default figure size 5 by 5 inches with 200 dpi resolution.

For concept maps

https://cmaptools.en.uptodown.com/windows