47-722 - Mini 4 - 2005
> Seminar in Finance II
>> Preferences, Behavior, and Finance
>>> Menu:
[Home]
[Syllabus]
[Topics]
[Readings]
[Assignments]
[History]
>>>> Announcements
- It is a wrap
- The course is complete!
See the reading list, references, and history to see what we covered.
- See you back here when studying for those comp/qual exams!
- April 27 (Meeting 14!!)
- We can complete our discussion of risk-sharing problems
by chatting about the infinite-horizon case
- This should leave us some time to reflect more generally
about modeling preferences behavior. So put on
your "big-think" hats!
- Discussed the dynamic risk sharing problems and
had a brief discussion on ``philosophy of science'' as it seems
to apply to financial economics.
- April 25 (Meeting 13)
- Monday, we can take a look at risk sharing problems. The big beef
(and a legitimate one)
against behavioral finance is whether or not the behavior survives
aggregation. With Chew-Dekel risk preferences and Epstein-Zin time/risk
preferences, we can capture a rich set of behavior. Do the
interesting behavioral features disappear if we have multiple
individuals with differing preferences. One way to look at this
question is to look at 'risk sharing' problems -- that is characterizing
the Pareto frontier.
- The place to start is
Lucas and Stokey [a copy is on my door]. This is a hard
paper to read, so do not get lost in the details.
- You can also take a look at the Kan paper. That paper sets
things up in the recursive setting we have been using. However,
the paper is mostly notation.
- Two-period risk-sharing problem in KP/EZ setting
- April 20 (Meeting 12)
- We need to finish our discussion of Epstein Wang and
the effect uncertainty aversion can (cannot) have on equity premium
- Next up, we will look at Epstein Schneider and dynamic uncertainty
aversion
- The paper requirement is due 20 April My guess is that
everyone is running behind. Not a problem, send me a note and let me know what you are working on and when you plan to get things completed.
See: [Evaluation]
- We discussed Ambiguity and the equity premium, A quick
pass through Epstein-Wang, and looked at dynamic consistency
with ambiguity aversion
- April 18 (Meeting 11)
- We will dive into the uncertainty
aversion Gilboa-Schmeidler [posted on my door] framework.
- We can take a quick
look at the asset-pricing setting of Epstein and Wang.
(This is a hard paper to read and not worth burning a lot
of time on)
- The additional wrinkle we will need to deal with is "belief
updating" in this context. We may get to that on Monday,
but will likely get to this on next Wednesday
- The section in Backus-Routledge-Zin has a small overview
of ambiguity aversion that might make a good place to start.
- We covered background of Gilboa Schmeidler and applied
it to a simple static portfolio theory problem.
- April 13 (Meeting 10)
- We will also take a look at the
Barberis, Huang, and Santos model. This model delivers many
of the same features as the Habit model. Here, however, the
behavior is justified in terms of Prospect Theory
(Kahneman and Tversky 1979)
- Completed the discussion of BHS today
- April 11 (Meeting 9)
- we discuss the Campbell-Cochrane (1999) "Habit" model. This
is now quite a standard approach to generating the necessary
time variation in the equity premium.
- The paper builds on Backus and Zin (1994) who look at the pricing kernel in the bond market
- We will also take a look (time permitting) at the
Barberis, Huang, and Santos model. This model delivers many
of the same features as the Habit model. Here, however, the
behavior is justified in terms of Prospect Theory
(Kahneman and Tversky 1979)
- Completed our discussion of Cambell Cochrane
- April 6 (Meeting 8)
- We will finish the Routledge Zin GDA paper
- Next we discuss the Campbell-Cochrane (1999) "Habit" model. This
is now quite a standard approach to generating the necessary
time variation in the equity premium.
- The paper builds on Backus and Zin (1994) who look at the pricing kernel in the bond market
- [My guess is that we will not get all the way through the Campbell
Cochran paper on Wednesday -- no worries]
- We finished discussion of Routledge and Zin and started in
on the Campbell Cochrane Habit model
- April 4 (meeting 7 of 14!)
- Reminder: Daylight Savings Time starts Sunday Apr 3 -- Spring those clocks forward one hour
- Melino and Yang (2003): We will look at their characterization
of the Mehra Prescott kernel
- Briefly, we can finish the discussion of the Melino Yang model
(and discuss the Gordon/St-Amour paper in passing)
- change in schedule Next we will talk about
Routledge and Zin (this is in the EZ class of preferences, so it
is more natural to slot this in here)
- I am not sure if we can fit this all in, so April 6 may
be a day for catching up
- We constructed the Melino-Yang pricing kernel for the
Mehra Prescott economy and started the discussion on Routledge/Zin
- March 30 (Meeting 6)
- Next we will apply the Epstein-Zin pricing kernel to asset pricing puzzles. We start
by looking at the pricing kernel in the Mehra-Prescott economy. This is constructed in
Melino and Yang (2003).
- Also for background, take a quick look at Table 3 in Epstein-Zin 1990. This has equity premium in a simple i.i.d. calibration
- For background, you might want to take a look at Gordon and St. Amour (2000). This
paper starts with standard time-additive expected utility with a time-subscripted coefficient of
risk aversion
- We spent most of the day talking about (a) how to
solve EZ and KP models (solve for price/dividend ration, and (b)
the equity premium in the i.i.d economy. (We ended up not getting
too deep into the Melino-Yang paper.)
- March 28 (Meeting 5)
- We will look at applications of recursive utility. In particular, we will look
at the representative-agent asset pricing model
- Look at Epstein-Zin (1989) Sections 5 and 6.
- If you are interested in the empirical implementation, look at Epstein/Zin (2001)
- Assignments
- Assignment 2 is posted. See [Assignments]
- On Assignment 1, do not get stuck on Q2(c). I can see how
to solve it with WRS but using SWS is odd. (If you solve it, let
me know)
- We looked at the consumption/savings/portfolio problem in Epstein-Zin
and constructed the pricing kernel (for the specific case of Kreps-Porteus)
- March 23 (Meeting 4)
- The topic for today is time and risk. The plan is to discuss what properties we might
want in preferences that involve intertemporal choice (e.g., time consistency) and
what sort of preferences deliver them.
- My suggestion for what you might want to read (I will try to update this later today
after I have put some notes together)
- Adding time/risk is a lot of notation. Try not to get lost
and focus on the concepts. We will do a few examples to see
non-time consistent and time-consistent preferences.
- Johnsen and Donaldson (1985) -- what are the definitions / issues
- Kreps Porteus (1979) -- Read introduction and Section 3
- [If you get a chance, Epstein Zin 1989 -- Focus on Section 3]
- [I will borrow an example from Selden (1978) -- but no need to read this one closely]
- [Koopmans -- The basic stuff on recursive utility under certain consumption plans]
- We looked at examples and definitions of time-consistent planning. The main result
is that time-consistent planning requires utility to be recursive. Next day, we look at specific
examples in the context of a representative agent asset pricing context
- March 21 (Meeting 3)
- We will spend some more time exploring the properties of non-expected utility. In particular, focus on a two period portfolio problem
- We will then turn to dynamic preferences. Dynamics introduces a new property of preferences: time consistent planning. For that, let's start with the Jonson and Donaldson paper.
- Assignment 1 is posted. See [Assignments]
- (A few typos fixed in version posted March 17 - 9.26pm)
- We looked at: Using Chew-Dekel utility functions. Focused on indifference
cures, first-order risk aversion, portfolio problem.
- March 16 (Meeting 2)
- A copy of the Chew (1989) paper is on my door.
- You might also look at "Risk" Section (pages 8-15) in
Backus, Routledge, and Zin
- We looked at: ''betweenness'' property and the axioms
that lie in the middle of ''betweenness'' (preferences that are linear-in-probabilities) and the strong-substitution independence axiom of expected utility
- March 8 (Meeting 1)
- First Meeting in room 318
- Course overview and axioms for v.n.m Expected Utility
>>>> Syllabus
Updated at: 08.09.2005 14:24
© Routledge 2005