For every dollar less of taxes, would you get one dollar of additional post-tax income?
No, because I might get a lower salary.

—

What if the government dictated that rent in my area was not to increase more than 1% a year?
My rent would probably be cheaper than rent in other areas.

Isn’t that great?
Well, not if the house falls into disrepair because the landlord find it not worthwhile to maintain the property after a few years.

—

What if the government raised taxes on companies?
Companies will pay more taxes.

Isn’t this a good thing?
Well, things might get pricier too, as those new costs are passed on.

—

Cross-elasticity is a very broad phenomena. Any profit-generating activity that involves multiple inputs and outputs has many ways of adjusting those inputs and outputs to maximize value (be it profit or consumption), and when prices are changed for one input or output it will affect the demand / supply of the other inputs and outputs in some way that can be difficult to anticipate.

To assume that the only change would be in the adjusted input / output is naive.

Related video: http://www.youtube.com/watch?v=9b-w7GOTqSY

Unlike computer games, there is nothing that guarantees that you are in fact playing a board game correctly. People learn to apply a set of correctness checks to the game – it can’t be that the first mover always wins, it can’t be that certain pieces are always ignored, and it can’t be that losing a small part of the game means you would never catch up. Whenever a game seems bad, the first thing that occurs is always that you’ve read the manual wrongly, and the response to that is to figure out some way to play differently so that the problem does not occur.

I’ve always found good social games to contain an equalizing dynamic, where those who are behind are provided with certain advantages that help them to catch up, whereas those that are ahead have to be wary of defending their lead. This is related to the fact that games differ greatly in the shape of their learning curves – in games like Go, with very steep learning curves, the chances of a novice beating an expert are pretty much zero, and there is pretty much no equalizer built in.

Demand:

• What percent of Singaporean household income is spent on housing? How regressive is this? How do we account for purchases using retirement funds (CPF)?
• How does the demand curve for living space depend on household size and demographic group?
• Do school districts have an observable influence? What about expat areas?

Supply:

• What proportion of the cost of an apartment due to the land?
• How linear is the dependency between apartment value and apartment size?
• How much cheaper are apartments with longer commutes?
• How big is the effect of regulatory limits on living space to land ratios?
• Does it make sense for the government to manage house prices because of their role as household collateral in a fractional reserve system? Are there significant interactions with monetary policy?
• How much monopoly profit does the local building industry enjoy?

I haven’t gone very far in the book, but it definitely seems very promising. The central premise that I can grasp is that it is fruitful to consider the quality of a model strictly in light of the decisions that the model is being used to make. This can be contrasted with typical surrogates like the root mean squared error or mutual information – those measures for model quality are largely derived using definitions of convergence towards total knowledge, and attempting to reach the state of total knowledge is often futile in reality.

In building practical models with practical approximations, choices in the early steps of an analysis often contain implicit knowledge about what the output from those steps is to be used for. To carry this process to its logical conclusion would be what I imagine approximate dynamic programming accomplishes – this way of dealing with ignorance trades off uncertainties in such way as to minimize the impact they have on decision quality, as measured by what you actually care about for the specific problem.

The advice is to structure the meeting like this:

1. Small talk and being human
2. Brief self introduction
3. Understand the client – ask questions and understand needs
• If no fit, end the meeting
4. Discuss price range
• If no fit, end the meeting
5. Find out more about how the purchasing decision gets made / what support or information the prospect needs to make the decision
• Try to get the process for this next step going
6. Schedule a concrete followup

I think that this is an excellent structure for the people who are networking – i.e., when the product is yourself. How would you sell yourself if you were going to follow these steps?

What makes a food, or a person, have a high fatigue point? Mere familiarity and agreement, or something more?

The better a question is, the more it helps to organize the underlying corpus of data, and help to turn a body of facts into knowledge. The person who is in actual possession of the facts isn’t the best person to synthesize it, simply because there is too much hindsight bias – if you have known something for too long, you lose the ability to differentiate the simple from the complex, and when you are unable to shun complex analyses in favor of simple ones you lose your sense of direction when engaging in critical reductionist analysis.

On the other hand, if you are able to, through interacting with someone intelligent but ignorant about the specific topic, communicate the entirety of your knowledge in a pithy way, then you know that you really do get it.

Without any form of quality control, it really isn’t that difficult to keep writing. There is a precedent for encouraging writing regardless of quality – Nanowrimo. Nanowrimo is an activity held every November, which aims to, for whatever reason, encourage the writing of 50,000 words within the month. I will continue work on my weak derivative of this activity – one blog post a day.