Finance 2 Investors and Markets Practical stuff On

Finance 2: Investors and Markets Practical stuff • On Wed. May 27, there will be regular tutorials. You can test yourself w/ some pop-quizzes from the course homepage. • On Wed. June 10, there are lectures from 13. 15 in [? ? ? ] The plan for today Mon. May 18, 2009: • Remarks on Sharpe’s Ch. 5. • Ch. 6 w/ disagreement over prob’s.

Cases 9 -10 Recall: Mario and Hue work for MFC, HCF. Points: • If you already have large exposure to something you should under-represent that in your financial portfolio to achieve diversification. Don’t put you pension savings in your own company. • The ”real” market portfolio is hard to observe; not just financial assets. Roll’s critique. • John Y. Campbell on "Household Finance".

Case 11 Recall: Salaries vs. collateral. Junior can’t borrow. Points • Investors with long horizons want to overrepresent stocks if human capital is safe. The latter is important; beware of spurious applications of the law of large numbers. W 5 Exercise. • The inability (even for one party) to make strong, binding commitments can hurt everybody. Bankruptcy protection isn’t. It seems possible to spin this both as “more” or “less” regulation is needed.

Chapter 6 Disagreement over probabilities: A main driver investment or: speculation, betting. Active managers remark. Index fund premises.

Intuition: If I think an asset is over-valued and you think it’s under-valued, then I want to sell and you want to buy. Large effect on our portfolios. Not so large on prices. Or: Prices are averages so the law of large numbers apply.

When Sharpe writes prices reflect information he means something like: - A market-beta relation, e. g. CAPM, holds to a reasonably good approximation. SML in short. - Not a lot of portfolios above the CML on which the market portfolio lies by def. - Individuals’ portfolios (well) below CML and scttered in SML-space.

Chapter 6 Cases #14: Setting the stage w/ Hue and Mario #15: Many independent predictions; vox populi. #16: Many are dead wrong, a few are right. Detectable gains. W 5 Exercise. #17: Wrong & noise (”bias”) vs. right: Harder to see gains. #18: Different prediction accuracies: Quite hard to spot.

Disagreement or uncertainty is likely to happen over expected values rather than variances. W 5 Exercise. Can we find a way to tilt probabilities in APSIM to achieve this? Less risk-averse investors many be willing to pay more to learn true parameters. (Bas, Roon, de Werker; Hand-In #2? )