Computational Investing Part I (Coursera) - Week 6

Week 6

Module 1

1.1

  • How to assess an event study (with a market simulator)
  • Event study allows you to test hypotheses:
  • What if Market Bollinger value is high
    • Look at equity and market to see if any discrepencies.
  • But stock Bollinger value is low?
  • Buy signal?
  • Simulator lets you operationalize
  • Transition from hypothesis to strategy
    • Buy/sell
    • How long to hold?
    • How much to allocate to each bet?

Module 2

2.1

  • Fundamental law of active portfolio management.
  • Warren Buffett quote: "Wide diversification is only required when investors do not understand what they are doing."
  • Thought experiment:
  • Stock trade = bet
  • Coin flip = bet
  • Uncertainty = beta
  • Coin bias = Alpha
    • 51% heads
  • Reward/Risk measures:
  • Possibility to lose everything
  • Standard deviations of returns
  • Alpha is lower you should can make more bets?
  • --Opinion-- didn't really follow this lesson all that well.

2.2

  • Active portfolio "investors" are not considered investors in Warren Buffett's eyes.
  • Jim Simons and Warren Buffet offer about the same performance
  • Buffet -- 54% in just 3 stocks (2010)
  • Simon trades actively -- 100ks per day
  • Algorithmic trades can "scale"
  • The two can be related with "the fundamental law"

Module 3

3.1

  • Covers CAPM.
  • Understanding beta and alpha of a portfolio
  • Didn't follow this so well. Really not interested.

3.2

  • Active Portfolio Management recommended again
  • More on using CAPM to build a portfolio (shorting companies, "longing" companies)
  • Hedge fund stuff. Not particularly interesting to me at this stage.