How to Build an Economic Model in Your Spare Time

1
How to Build an Economic Model in Your Spare Time

• Advise to Graduate Students from Hal Varian.
  (December 1994) from Passion and Craft
  Economists at Work edited by Micahael Szenberg,
  University of Michigan Press, 1997.

2
Getting Ideas

• Tricky part is to get a good idea
• Come up with many and throw out the bad ones.
• Source of good ideas?
• Journals not very good for original ideas. Get
  techniques and insights from journals, not ideas
• Look outside of journals (newspapers, magazines,
  converstaions, tv, radio.)

3
Assuming you have an ideais it worth pursuing?

• Try to phrase your idea in a way non-economists
  can understand. (if you cant do this its probably
  not a good idea). Remember Marshall.

4
Marshalls advice

• 1) Use mathematics as a shorthand language rather
  than as an engine of inquiry
• 2) keep to them until you have done.
• 3) Translate into English
• 4) illustrate by examples that are important in
  real life
• 5) Burn the mathematics
• 6) if you cant succeed in 4, burn 3.

5
is it worth pursuing?

• Is it interesting?
• Try it out on a few people and see if they think
  it is interesting.
• Remember working on an idea has an opportunity
  cost
• Does it generate an insight an Ah, oh, I see
  or a So that is why

6
Delay the literature search, because

• You need the practice of developing a model, even
  if you end up reproducing something already in
  the literature.
• You might come up with a different approach.
• Your ideas need time to incubate, then your ideas
  can interact with those in the literature.

7
Building the Model

• Basic format usually the same
• There are economic agents
• They have choices to make in order to achieve
  their objectives
• Choices are made subject to various constraints
• Something adjusts to make the choices consistent.

8
Questions as a Plan of Attack

• Who are the people making the choices?
• What are their objectives?
• What are their constraints?
• Who do they interact?
• What adjusts if they choices arent mutually
  consistent?

9
Building the Model, Stage 2 work some examples

• Start with the simplest possible example. One
  period, two goods, two people, linear
  utility---whatever it takes to get to occams
  razor.
• Work another example.
• See what is in common
• Write down your model. KISS
• Write down the simplest possible model and see if
  it still exhibits some interesting behavior.
• Keep at it till it gets very simple.

10
Advice from Einstein

• Everything should be as simple as possible but no
  simpler. A model is supposed to reveal the
  essence of what is going on. Your model should be
  reduced to just those pieces that are required to
  make it work.

11
Generalizing the Model

• Your model is most likely a special case of one
  you have learned (consumer choice, producer
  choice, general equilibrium, etc.(
• Apply the results concerning the general model to
  your specific model.
• Make mistakes and try again.
• Chip away until the model is revealed.

12
Search the literature

• Ask around as well as looking on the internet or
  Journal indices
• If someone else started with your idea and
  carried it farther, you want to see what you
  missed.
• Maybe you have an insight they didnt or yours is
  more general.

13
If your idea is not in the literature

• Maybe you made a mistake
• Check with your advisor
• If you have made your analysis as simple as
  possible
• It is less likely to contain an error
• An errors that remain will be easier to find.

14
You may be too close to the problem

• Get others opinions
• Go back to basics
• What is the fundamental idea
• What are you trying to do

15
Give a seminar

• You get immediate feedback from your audience
• A seminar is a good way to get some attention
  that might otherwise be hard to get

16
Parts to a Seminar

• Introduction dont have one- or short and right
  down to business.
• Content-your audience will only remember the
  first 20 minutes of what you say so make sure its
  good.
• Conclusion summarize what you did and why its
  important

17
Media

• No more than one or two for intro and conclusion
• Big slides with limited information large
  type..dont try to say too much.

18
The content

• Establish your credibility.
• Very early go into GREAT DETAIL in the
  presentation of your first result a theorema
  regressiondiagram---whatever.
• Spell out each aspect of your result in
  excruciating detail so no one can possibly
  misunderstand.
• Stay on taskDont get sidetracked by a question.
• Stay in control. Dont let them sleep but dont
  get them too excited
• Then go on to generalizations and elaborations.

19
After the presentation

• Jot down notes was it difficult for people to
  understand?
• How did you finally get it acrossif you did.
• What suggestions were made?
• What references were given or suggested?

20
Planning the paper

• Have an outline.
• Have a file for references
• Have a file for ideas
• When you have filled in the outline, then use it
  to expand to the paper. You will probably have to
  reorganize. Keep checking the notes for ideas to
  see if you have included them or if they make
  more sense.
• Put the draft away for a couple of weeks
• Read it and critisize as though it were someone
  elses work. Keep all revision versions. Put it
  away again and repeat.

21
Structure of the paper

• Introduction is most important section. Should
  be brief. Should be clear. Everyone should
  understand it. It should grab the reader.
• The rest of your paper should be like your
  seminar presentation.