Posts

Comments

Comment by DogHead on Risk Premiums vs Prediction Markets · 2021-07-29T15:08:29.335Z · LW · GW

Assuming Sharpe tops out at 0.5 seems wrong. Over the last 5 years, Bitcoin has had a Sharpe ratio of 1.6 the Wilshire 5000 has had a Sharpe of 1.04.

Comment by DogHead on Care and demandingness · 2021-03-09T02:34:00.573Z · LW · GW

Demandingness can apply to objective claims as well. “There is a storm coming that will destroy everything”. The fact that stuff exists now makes such a storm less likely; not impossible but certainly evidence against a storm of that severity. Existence can’t be so hard as to preclude life, we really did catch a break. The anthropic principle seems directly analogous to what you are getting at.

 

Many people believe ethics is situated. Just as knowledge is known by someone, moral acts are done by someone and towards some end. The fact that, on the margin today, you can pay a charity to help save lives does not change the need at its base level. It is kind of weird to care as much about people you will never meet as your friends and family or even your mere neighbors. Current mainstream philosophy may agree with you but the public at large remains unconvinced.
 

Budgetary demands for all available resources really are different in kind than budgetary demands to be included for consideration at all. Something that can claim all available resources closes off all other values being traded against. If human lives are the only thing that matters, better not donate to the EFF. Is it unethical to care about an open internet? Being underfunded is better than being not funded at all. Once something garners no resources, whether that is time or capital, the capacity dies outright.

 

Good consequentialism may want to consider how things have actually played out historically. Prosperity has not come by making man a more generous animal. On the contrary modern technology seems to require a market for toys or novelty. Cell phones for jet setting egotists in the 80’s have done more to liberate today’s global south than any amount of charity dollars. It is even arguable if an agrarian society was to maximize on alms for all, they may indefinitely forestall an industrial revolution, immiserating the generations to come for present comforts.

Comment by DogHead on Asset Prices Consistently Violate Efficient Market Hypothesis · 2020-07-31T21:02:14.212Z · LW · GW

I don’t think this summary does a sufficient job of capturing the nuance of the argument being put forth. It also might be giving the impression that the critique is based on a misunderstanding of EMH, which his first paragraph well states. The author is directly claiming there is a clear mis-pricing in assets that offer a greater than total market risk-adjusted return. It is one thing to claim he is talking his book, which he is, it is quite another to ascribe ignorance to the view articulated without reckoning with any of the arguments.

Momentum as a factor is hard to explain under EMH because if the current price incorporates all known information, how can the time series of price information contain a trade-able edge?

The example of HKL is put forward, which is a holding company for prime real estate in Hong Kong and Singapore. The stock trades at valuation of only 25% of the most recent market value of their net real estate holdings. There are many possible reasons for this, and it is fairly common for holding companies to trade at a discount to their net assets. The author argues this is due to an anchoring type effect. The shares of HKL are traded on an exchange among other stocks. The salient benchmark being the rates of return of the rest of the equities in the market. To match the higher yields available from other equities HKL must trade at a discount to book.

The underlying real estate assets are being valued differently on the open market. Here a 4% yield looks great if the salient benchmark is fixed income, which is reasonable because of the high quality of the cash flow. If both of these are true the same asset is being valued differently based on the market it is being traded in. This not only violates EMH, but also the law of one price. This is an arbitrage opportunity but the onus is on a market participant to articulate and execute the trade that captures this opportunity.

Crucially he is not arguing the market is not incorporating public knowledge efficiently. The argument states there are principle-agent problems writ large in financial markets the allow for persistent mis-pricing of assets. To capture these opportunities you don’t need better knowledge but a better structure than the average market participant.

Whether EMH is true or not, it seems to get a lot of free credit as a theory. It proves so little and commits its adherents to so much. I think its popularity may lie more in its beauty than in its truth.