Monday, January 26, 2015

One of two children is a boy, how likely is it that the other is a boy?

One-half. Not one-third; one-half.

Since I work with probability and statistics, I sometimes endure someone trying to 'teach" me that it's one-third. Because I've tired of explaining why 1/3 is the wrong answer, one person at a time, I made a video:



Essentially the problem is foundational: people who get the wrong answer do so because they haven't learned to think in terms of states, events, and their probability implications.

Saturday, January 24, 2015

The easier it is to graduate college, the bigger the advantage of graduating


The easier it is to graduate college, the bigger the advantage of graduating. Obvious, no?

Wait, what?

"No, no, no, I hear you say. The harder it is to graduate college, the bigger the advantage of graduating." This sounds reasonable: if graduating college is very hard, then being a graduate signals high ability. But alas, that seems to forget the effect on the people who don't graduate.

First let's have a point of agreement: the easier it is to graduate college, the worse the average graduate will be. (Keeping the population constant, of course.) Any reasoning, or model, that doesn't capture this effect will be prima facie wrong.

Now to the point of disagreement: that the easier it is to graduate college, the more important it is to graduate.

Let's consider a job market choosing between a college graduate $G$ and a non-graduate $NG$, based solely on that fact. Then, what the job market should care about should be the average "quality" of these two groups of people.

(We'll assume that there's some "quality" $q$ that determines success in college that is correlated with what the job market wants.)

Let us study, then, the behavior of $E[q|G]$ relative to $E[q|NG]$ as a function of a threshold quality needed for graduation $T$. The definition of $T$ is that all members of the population with $q \ge T$ graduate college and all others do not. This is obviously an approximation.

First, let's consider a simple population, with quality uniformly distributed on a finite support (this is called a Hotelling model), which without loss of generality and significant gain in simplicity we assume to be the interval $[0,1]$.

We'll use as a measure of the advantage of college the ratio of expected quality,
\[ \frac{E[q | G]}{E[q | NG]} = \frac{E[q|q\ge T]}{E[q | q < T]} = \frac{(1+T)/2}{T/2}
\]
A few days ago I plotted this function while microwaving my lunch:


The chart also shows the average quality of a college graduate, and it does decrease as the fraction of people graduating (i.e. the ease of graduating) increases.

Later that day, having another food-related wait, I decided to do the same for a standard Normal distribution, $q \sim N(0,1)$.
\[
 \frac{E[q | G]}{E[q | NG]} = \frac{\frac{\frac{1}{\sqrt{2 \pi}} \,\int_{T}^{+\infty}  q \, e^{-q^2/2} dq}{1 - \Phi(T)}}{\frac{\frac{1}{\sqrt{2 \pi}} \, \int_{-\infty}^{T} q \, e^{-q^2/2} dq}{\Phi(T)}}
= \frac{\frac{e^{T^2/2}}{1 - \Phi(T)}}{\frac{- e^{T^2/2}}{\Phi(T)}} = - \frac{\Phi(T)}{1 - \Phi(T)}
\]
Note that this is always a negative number, an artifact of having the distribution centered at zero, so that $E[q|q < T] \le 0$. Since it could easily be shifted into positive values simply by adding a positive mean to the distribution (and making the formula above a bit unwieldy), it's irrelevant, only the relative positions on the plot matter. If you're uncomfortable with that, be my guest, add a $\mu >> 0$ to the formula and make your own plot.

Here's the plot for the normal distribution, with $T$ in the scale of standard deviations:


One simple way to see how this works is the following: if college is really easy to graduate (which includes access to college and financing, by the way), only very unmotivated people will not graduate college, as long as the job market expect you to.

Perhaps instead of getting more people into college, society should work towards not having an expectation of a college degree.

Friday, January 9, 2015

Three lessons from teaching MBAs in 2014


Use longer, content-heavier handouts; integrate local and up-to-date content; and show numbers and math.


Change 1: Longer and content-heavier handouts

The only significant complaint from previous cohorts was regarding the lack of a textbook. I post a selection of materials to the course support intranet (consultancy reports, managerial articles, academic papers, book chapters), but a few students always remark on the lack of a unifying text for the class.

(There's no unifying text because -- in my never humble opinion -- most Consumer Behavior textbooks are written from a consumer psychology point of view, while I prefer a more marketing engineering point of view.)

Taking that into consideration, I made longer, denser handouts, each like a book chapter rather than just support for in-class activities. The class is participant-centered, with minimal lecturing, so these longer handouts help students feel that they have a coherent framework to fall back on.

Handouts changed from a median size of four pages of mostly diagrams, in 2012, to a median size of eighteen pages of text, diagrams, and numbers, in 2014. (Just a reminder, since there's some confusion about it, that handouts and slides serve different purposes.)


Change 2: More local content

I used local content in most class sessions: local products, merchandizing from local retailers, and examples from local advertising. In particular, using outdoors from around the campus allowed students to recognize their location, for a little a-ha moment that improves mood.

The main advantage of local content is student familiarity with it. Examples are more effective when students don't have to learn new brands, new product categories, and other regional differences. A disadvantage is additional preparation work, but that work also signals to the students the instructor's commitment to the class.

A secondary advantage of local content is as evidence of instructor competence. Local content, and up-to-date content, requires confidence, ability, and practice. For this reason alone, it's worth the additional work, even if old or foreign examples would be equally good for teaching.


Change 3: More numerical content

The rise of analytics is a highly visible trend in marketing; marketing courses are therefore increasingly quantitative. Still, most Consumer Behavior courses shy away from math.

Our course was different: there were plenty of numbers and models. I did most of the work, not the students, since the objective was not to teach them analytics; but I did do the work, so the students were shown modern marketing techniques rather than a lot of hand-waving.

For example, to illustrate the effects of memory on different types of advertising timing, I used a computer simulation of a learning model: instead of rules-of-thumb for media planning, students saw how learning and forgetting rates change the effectiveness of blitz versus pulsing media timing.

(References to technical materials were provided for students wishing to learn more, of course.)


Results

Despite objectively covering more material than before and using harder assessment measures, student grades were higher. In other words, these changes achieved their primary objective: students learned more material and learned it better.

Class dynamics were better than before, though they were pretty good in previous years. When I pick up my teaching evals in 2016 (they're on paper), I'll know whether I kept my top-5 ranking from 2012.

Addendum:  In short events since the MBA class, I replicated these three changes, yielding performance improvements along all dimensions: participant learning (as measured by in-event exercises), participant experience (as measured by client-run event evaluations), follow up contact with the participants, and word-of-mouth.

Friday, September 5, 2014

On hiatus again

(Gee, really?)

Blogging to resume when I can figure out the balance between the enjoyment of sharing valuable knowledge and my MBA drive to monetize anything of value I do.

Also, work.

Saturday, April 5, 2014

How to succeed as a popularizer of technical material without knowing anything

The problem with financial journalism: journalists
[T]his is the formula for selling middlebrow “popular science” books. Flatter the reader that he is a smart monkey privy to the secrets of the universe; so much smarter and better informed than those people who actually work in the field. Popular science writers don’t sell actual popularizations of science like they did when Asimov used to write them; modern popular science writers now sell smugness. Modern upper middle class over-educated people love smugness, and use it as a sort of barter currency in social interactions with the fellow enlightened.
A perfect description of the genre and its audience. Applies equally to all technical material: science, technology, engineering, economics, business, management.

I continue to ask my kinetic energy question to people who want to "talk science" to me, and still find that most of them can't answer it. And don't get me started on "analytics" or "big data" advocates (as opposed to practitioners) who don't understand -- sometimes know -- Bayes's formula.

Thursday, February 6, 2014

Six rules for my financial bliss

Really there's only one rule: make more money than you spend. Much more, if possible.

These rules work for me and are not to be construed as investment advice. They based on three components: basic knowledge of finance, good understanding of economics, and expertise in marketing (in other words, the art of making others spend money – knowing it is like being inoculated against it). Some people might get ideas for their own rules from these, so without guarantees of any kind, here they are:

1. Investment assets should produce income (yes, they could produce a capital gain instead, but I'm not a fan of counting on capital gains as a life strategy); an "asset" that produces a need for income (for example, a house bought with a mortgage) is more correctly defined as a liability. The best investment asset I can acquire is expanding my skill set, and the interwebs have made that almost free. (I'm weary of leverage, even for income-producing assets.)

2. There are many things I'd like to own that I don't really need. Using a total cost of ownership model, including the space and time cost, rather than simply looking at the price, and comparing with alternative uses for money and alternative sources of happiness, I generally stop myself from buying anything but consumables. For example: books, once one of my largest like-driven expenditures, I have all but stopped buying. I rely on libraries for most and have set a rule not to buy more than five books a month, avoiding paper books when possible.*

3. Exceptions to rule 2 are: small luxuries and a monthly "slush fund" of $100 for impulse purchases, with the proviso that they have to be physically small, require no maintenance, and enrich my life in the long term.

4. Proper maintenance and care, coupled with high-quality purchases to begin with, are key to asset longevity. Bespoke suits physically last longer and look stylish much longer than designer suits, so their yearly-amortized cost is much lower. (Is it noticeable that I aced all MBA accounting and finance classes, despite being in a marketing and strategy concentration?) I avoid any assets with planned obsolescence paths if I can and never buy anything for identity reasons. (I do buy some experience products; there's no real defense against those except no-exceptions thrift.)

5. When earnings increase, I feel no obligation to increase spending in fact fight the temptation to do so. (Note that I live a very comfortable life, with no privations; this rule would not apply to someone just muddling through.)

6. Long-term forecasts of economic variables are about as reliable as long-term weather forecasts. I trust estimates of growth and inflation for 2030 about as much as I trust today's rain and temperature forecast for San Francisco on Jan 3, 2030.

Yes, these rules are like the way to lose weight: control what you eat and exercise diligently. There's one extra rule, though, for people who are disappointed by the lack of a magic solution to financial problems:

7. If something sounds too good to be true, it probably is. Money-wise and otherwise.


-- -- -- --
* During three separate decluttering events in the last 8 years, the hardest part was getting rid of books: in the first one, I only got rid of outdated textbooks; in the second one I didn't touch the books at all; in the third – radical – one, I went from a little under 2000 books to just below 500, but took weeks doing that, while the selection of around three cubic meters of designer clothing for donation took but a couple of hours. Electronic and library books will spare me further trauma.

Friday, January 31, 2014

The books of January

The unusually cold Bay Area weather colluded with the short January days to give me a lot of time to read; here's the result. Books listed in order of acquisition; reading was interleaved and out of order, as usual.

1. Henry Petroski: To forgive design. (I'm a fan of Petroski's work, and have all his books.) In this book Petroski extends the study of engineering failure to include systemic components and human decision-making, getting into Donald Norman territory. For anyone interested in technology, engineering, and understanding the way things work and fail, this is a must-read.

2. Matthew Jackson: Social and economic networks. This is technically a work book, but I already know its contents, so I read it for pleasure. It's reasonable as a set of class notes for Jackson's course and MOOC, fairly technical for the general educated public.

3. Charles Gasparino: The sellout. Reads like a novel about the crash of 2008. Gasparino,  a journalist, writes better than former trader Michael Lewis, with better information and broader vision (because Lewis was in "the trenches" where vision is necessarily limited), and without a chip on his shoulder against the financial system. Read in two sittings in two consecutive evenings. Take that, tell-lies-vision.

4. Duff McDonald: The firm: the story of McKinsey. I won't comment on this book, other than to say I highlighted a large number of passages.

5. Richard Hammond: Or is that just me? (Library book; doing my part to slow down economic growth.) The self-described short bloke from Top Gear writes about the making of some of his shows and some personal stories. It's not bad per se, but it's not funny or insightful, like the books by the louder, bigger Top Gear co-host Jeremy Clarkson. I started skimming, then skipping ahead, and eventually gave up in the middle. Sorry, Hamster.

6. Charles Gasparino: Circle of friends. (Library book.) Again reads like a novel, and I read it in one sitting on the weekend. If you like financial and governmental shenanigans, read it: it describes the witch hunt criminal investigation of insider trading in the late 2000s. As one of those involved points out, it's funny how insider trading is touted as a heinous crime, and people serve time for it, but it doesn't make people poor, like bubbles, which have produced zero arrests.

7. Charles Gasparino: Blood on the street. (Library book.) Unsurprisingly, it reads like a novel and I read it in one sitting on the weekend. The same weekend. (Bad weather weekend.) If you like financial and governmental shenanigans, read it: it's the story of the late-90s stock boom and bust, with an emphasis on the conflicts of interest between analysts and investment bankers in the same firms, while the various regulators looked away, then political opportunism in finding a few scapegoats. (This is my interpretation of the book, not how it's presented.)

8. Peter Mayle: Acquired tastes. Got it after reading this post at A Suitable Wardrobe. A fun description of some luxuries by a member of the upper middle class — which looks sober and thrifty compared to the lifestyles of the rich and shameless of the third Millenium. Also a infomercial in text for Provence (not that it needs one). Two examples. Laugh-out-loud, especially for anyone who has spent time in France outside Paris, ski resorts, and beaches.


During January indoor rowing, elliptical walking, commuting, shopping, cleaning and other dead times, I also heard a few audiobooks (fewer than usual, but I listen to a few interesting podcasts and they compete for the same aural time):

9. Clayton Christensen: How will you measure your life? Reflections on what is important in life; a little more philosophical than one would expect from a Hahvahd B.S. professor. Made me think hard about some of my life choices, which is all that can be asked of a book like this.

10. Oscar Wilde: The importance of being earnest (performance). Obviously a reread, and I've also watched a few live performances and movies based on it before. Wilde's writing is magnificent and the actors (not readers) do a great job. The play is available free on the interwebs.

11. David Weinberger: Too big to know. Only heard part of it, still about two-thirds to go, but it's not grabbing my attention, to say the least. Perhaps a harsh assessment, since I have a professional interest in the issues of the book. But in general I tend to like popular books on my area of technical expertise, so the problem is mostly with the book, not any professional elitism on my part.


As is my habit, I reread bits and pieces (sometimes large fractions of the book) of P.G. Wodehouse, Evelyn Waugh, Mark Twain, H.L. Mencken, Ian Fleming [James Bond], and Agatha Christie [Poirot] here and there. Even some Umberto Eco essays! I haven't read any new fiction, since I find most of what's on offer formulaic and preachy, often poorly written and scarcely researched as well.

(Someone pointed out that in my "Books of 2013" post I missed quite a few, like the haul from the SFPL shown in this photo or this one. That's because I didn't log them; the system isn't perfect.)




IN THE QUEUE:

In the reading queue for February (not counting work-related materials, which take about half of my "leisure" reading time), meaning books I already acquired but had no time to read in January:

a. Niall Ferguson: High financier: the lives and time of Siegmund Warburg. (SFPL book.) Got the library copy to reread - I had read a library copy in 2010. (Slowing down economic growth even then.) Returning books to the library recently, I saw it and checked it out on impulse. That a person would choose to reread the biography of a mid-century banker on impulse is review enough for the book (or the person!), but to be fair to Ferguson: unlike most narrative writers now, who pepper their prose with irrelevant distracting details, NF picks the interesting parts of the story and paints a picture of a sober, vaguely obsessive, cunning banker who lived an interesting life.

b. Burkholder, Grout, and Palisca: A history of Western music, 8th edition. (SFPL book.) I own the portuguese translation of the third edition and I'm considering buying the upcoming tenth edition. Since it's a textbook and therefore costs its weight in DeBeers diamonds, I decided to check this intermediate edition to get an idea of the changes. (Textbook publishers use unnecessary proliferation of editions to undermine the used book market, and the parts of Western music that I care the most about happened too long ago for their history to have changed much from edition to edition.)

c. Lawrence Freedman Strategy: a history. Books that take a very broad view of strategy tend to be either good (rarely great) or very bad. I saw his talk at Google and since the book was only $10 on Kindle I bought it on impulse. (Publishers: cheap books mean a lot more impulse sales; do your math.) The sample was not too bad, so it might not be one of those very bad books.

d & e. Peter Mayle: A year in Provence and Toujours Provence. (Two eBooks, plus an abridged audiobook version of the two.) Given how funny Acquired Tastes was, I decided to try a couple more; added the audiobook so I can enjoy it on assorted torture repetitive exercise machines.

f. Peter Mayle: French lessons: adventures with knife, fork, and corkscrew. (Audiobook; considering the companion Kindle book to try Amazon whispersync.) Another Mayle book for exercise occasions. I expect that these books will put some stress on the Paleo, but then again – to paraphrase my PhD advisor, a very wise man – life is too short to eat bad food.

g. Lucius Beebe: The provocative pen of Lucius Beebe, Esq. (SFPL book.) Beebe is a chronicler of San Francisco in the old days, and I like the history of this city. People who protest ongoing changes in the city seem to not understand its history. San Francisco has changed its character more often than any city I've ever known, and that's one of the exciting things about it.

h. Herb Caen: Only in San Francisco. (Berkeley Public Library book.) Another chronicler of San Francisco's past. I like the city, I really do. I just don't want to live in it; the suburbs are much nicer than the city, now. Caen calls San Francisco Baghdad-by-the-bay in an eponymous book; Beebe's description, older, is more like Paris in the West Coast, with Boston being London on the East Coast.

i. Burt Folsom: The myth of the robber barons. On the recommendation of Free The Animal, I watched the author's video, so I'll read the book to balance out all the Aaron Sorkin series (The West Wing, The Newsroom) and movies I have watched.

j. Steven Kotler and Peter Diamandis: Abundance. (Audiobook.) Strangely enough, I find listening to popular books about my areas of expertise a good way to relax after work (on top of a elliptical machine, an indoor rower, or a treadmill – that's an essential part; never during resistance training, all 11 minutes per week of it).

k. Erik Brynjolfsson and Andrew McAfee: The second machine age. Ok, this one is close to a work book, but it isn't technical (at least from the quick browse), so I'll count it as leisure reading. (My work books have phrases like "compact space," "support vector machine," and "Markov perfect equilibrium." Also, they cost their weight in Beluga caviar.)

(That's eleven books – plus one abridged audiobook and an art book, Architecture Now 9, which should arrive tomorrow (it did) – in the queue and February hasn't even started yet. I already pre-ordered a couple of books coming out in March. It never ends.)




HOW THIS IS POSSIBLE:

The secret to reading all these books is very little television. (Also, being a vaguely misanthropic bachelor with no children.) For example, arriving home at 6PM I brew a pot of tea, sit in a comfortable chair, and read until dinner at the civilized European hour of 10PM; many people I know vegetate in front of the tell-lies-vision for the same period. (Or have families.) Reading four hours a day, eight or more per day on weekends, even if half of it is work-related, means two or three books per week.

Here are a few recommendations for improving the reading experience:

A comfortable chair is of paramount importance. Not an ergonomic office chair; a club chair. A tall overstuffed leather chair invites the relaxed feeling necessary to enjoy a book; ergonomic office chairs demand efficiency, measurable return on time invested, results Now! An ottoman is a better foot rest than a La-z-boy-type contraption designed for bovine ingestion of tell-lies-vision.  Fireplace and labrador retriever are optional.

If I'm reading on an electronic device, an increasingly common situation, I put it in airplane mode, blocking out the World Wide Wasteland. For the same reason I never read with the television on, even if muted.

Soft warm light is very important, especially at night; there's the matter of intensity (too bright you'll feel like you're being interrogated, too dim and your eyes will soon be tired) and of warmth, or hue. Most fluorescent lights (and cheap LED lights) have too much blue and violet in them; get a "warmer" lightbulb, one with more yellow and red, especially if you read a lot at night. Or bounce a "cold" light off a warm-colored reflector.

I like to have soft background music, usually something from between mid-16th and early 18th centuries; the mathematical precision of Buxtehude or his pupils J.S. Bach and Handel can serve as a mental rest area when I pause to parse a thought. Friends tell me they use jazz, but I find it distracting.



Reading: it's a cheap hobby, an invaluable habit.