Skip to content

bvi1994/CoventureTakeHome

Repository files navigation

CoventureTakeHome

Conventure Take Home

This take home is for Conventure. Note: I don't have any experience or background in Finance. However, I did the best I can to learn the the concepts of finance to do this take home and try really hard to focus on the programming aspect of the take home.

Nevertheless, I have enjoyed this assignment. Learning the fundamentals of finance and getting reacquianted with data visualization libraries like Plot.ly was so much fun. Hopefully I will get to do more such projects in the future with Coventure!

If you have any questions or concerns or any bugs in the code, please contact me at [email protected]! Enjoy!

Part 1:

  1. Go to Yahoo Finance and download a csv file of daily historical data on AAPL from Jan 1, 2002-Jan 1, 2010.
  1. Build a moving average strategy that tests a 5-day and 20-day moving average crossover. The strategy should go long when the 5-day moving average crosses above the 20-day moving average and short when the 5-day moving average crosses below the 20-day moving average.

In the 'longorshort' column of the csv, the date as to whether to go 'long' or 'short' is given.

  1. Build another strategy that buys AAPL when it closes down 3 days in a row and then sells the stock 1, 5, and 10 days later.

Part 2: 4) Graph the equity curve of the strategies

  1. List each trade in the strategy
  1. What 3 metrics would you analyse to determine whether the strategy was good or bad (not including the overall PNL of the strategy)?
  • The total return in percentage
  • Volotality/Stablity of investment
  • Risk Adjust Returns (Example: Sharpe's ratio)
  1. How would you allocate capital between the two strategies?

I would focus more on allocating capital on the long-short sale since in terms of absolute growth, that seems to be more profitable than going through 3-days of downward prices.

About

Conventure Take Home

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published