Катя (inostranka) wrote,
Катя
inostranka

Wharton | 10/1 and 10/2

Analytical courses only on both day - decision making and uncertainty, statistics and the-put-to-sleep-er macroeconomics. Second year we have almost entirely electives, and I'll be sure to pick as few analytical courses as possible, going for management (strategic and people), policy (business and social), and communications, instead. Quite the opposite of the MIT days, when any "soft" subject was looked at with scorn...

Optimization
"If you can't make it, fake it" - the optimization professor explaining the rationale for Monte Carlo simulations.

Equity considerations when trying to optimize "total happiness" in things like student dormitory assignments, course lotteries, etc. The problem is that the solutions that maximize total happiness (e.g., the sum of desirabilities of a given assignment by a given individual) often achieve optimality at the expense of a few unlucky chaps. This is an interesting philosophical problem - it is easy to make most people happy at the expense of a slighted minority. Justifying equity - reducing total "happiness" to reduce the number of dramatically poor assignments - is difficult politically.

Statistics
"Always do a pilot study, except that in the real world you will be told that there's no time or money to do one." This sounds so familiar... At least I hope that if they do create a pilot study, they do not end up actually using its results instead of using them to improve the real study. In the software world, the prototype so often becomes the end product.

In the UK they managed to guarantee a 100% response rate to a study on traffic signs. Cops were pulling people over to conduct the study and the poor folks were all too happy to fill out the questionnaire once they realized they weren't being fined for speeding. Wouldn't fly in the US, though.

Macroeconomics
What are the sources of long term GDP growth? In the US in the post war years - 25% due to investment (improved market liquidity and lower interest rates), 37% due to increase in the labor force (more women, later retirement age, population increase), and 37% due to improvements in skills and technology. How the heck do they come up with these numbers? Well, I do have a book that gives some justifications, but, basically, you'll have to trust my professor on this.

Another installment of Optimization
Discussed a case study of the Exxon and Mobil merger (using the Monte Carlo simulation to predict future oil prices). Apparently, they decided to merge in the summer of 1998, with oil prices having come down from $27/barrel to $17/barrel and heading lower still. Such low prices dictated mergers to create cost savings. Of course, prices have about tripled since then, but, nonetheless, the combined ExxonMobil is doing quite well. One of my classmates has been working in Exxon for the past 7 years and gave us the inside scoop.

I'll miss our Israeli professor (we have only one more class left, as this is a half-credit course). He smokes for 4 cigarettes during the break, drinks two cups of black coffee, and is so over-the-top hyper that it's contagious, even though the subject matter is, well, quite bland. And, as he himself admits, it's the tangents that are really important in this class, and, I might add, in all others - and we get off on plenty of those.

Flight Back
My favorite office - a plane seat. Where else can you be served food and drink and have an amazing and constantly changing view out the window? That, and no interruptions from cell phones, meetings, or emails? On the London flights there are movies, too, which kick in right after the battery on the laptop dies, but even the Philly flights are worth it.
Tags: wharton
Subscribe
  • Post a new comment

    Error

    Anonymous comments are disabled in this journal

    default userpic

    Your reply will be screened

    Your IP address will be recorded 

  • 15 comments