Lecture 7: Risk Preferences I

MIT 14.13 Psychology and Economics, Spring 2020
Instructor: Prof. Frank Schilbach
View the complete course: ocw.mit.edu/14...
KZbin Playlist: • MIT 14.13 Psychology a...
In this video, Prof. Schilbach describes how economics looks at risk preferences, that is, choices involving risk. Specifically, he covers the topics of risk aversion, expected utility, absurd implications, and small vs. large-scale risk aversion.
License: Creative Commons BY-NC-SA
More information at ocw.mit.edu/terms
More courses at ocw.mit.edu
Support OCW at ow.ly/a1If50zVRlQ
We encourage constructive comments and discussion on OCW’s KZbin and other social media channels. Personal attacks, hate speech, trolling, and inappropriate comments are not allowed and may be removed. More details at ocw.mit.edu/co....

[@devonrd]

For those wondering: at 31:23 the answer should be: ACCCEPT (not reject as shown in the slide)

[@theirishpizzaguy6663]

lecture starts 27:30

[@endgamefond]

I took your advice.

[@meteorxmoon775]

人们之所以对大金额的风险比较无感，是因为在生活环境中，人们单位时间的收入95%是落在小差别范围内的，也就是说大金额对人们意味着更长时间的收入加总，所以在考虑人们对大金额的在乎程度【损失厌恶】，还要考虑到人们如果通过工作获得大金额的时间是很长的，所以应该还要考虑到时间的长短【时间偏好】，将获得大金额的时间长短和时间偏好引入模型，建模，就会得到人们对大金额无感。人们在直觉的层面上已经加了这一层模型。

[@kagakai7729]

MIT professors could make wool carpeting interesting.

[@solarestone]

*Wool* carpeting you say? 👀

[@chuanqisun]

KZbin comments could make any professor interesting.

[@juanriojas1926]

I married a Latina. I have a high-risk preference..

[@Marteenez_]

At 38:00 he describes linearity in probability, but I don't understand his explanation. He says that if something has double the chance of happening we do not double it, which is pretty standard, given the join probability formula. However multiplication IS a linear transformation. Therefore, as probability is nonlinear, as shown by a pdf, shouldn't that mean that utility should be nonlinear in probability?

[@arkarupbasumallik]

No, that is not what he meant. One can explain this using an example. Suppose you have a probability=0.3 of having a wealth of 100 and a probability of 0.7 of having a wealth of 200. Suppose also that u(100)=3 and u(200)=5. Then the initial expected utility, EU=(0.3)u(100)+(0.7)u(200)=(0.3)(3)+(0.7)(5)=0.9+3.5=4.4 Now, suppose that the probability of having a wealth of 100 doubles from 0.3 to 0.6 and consequently the probability of having 200 as wealth changes to 1-0.6=0.4 but the u(100) and u(200) does not change. That is, the instantaneous utilities, alias Bernoulli's utility functions does not depend on probability of states. But the expected utility (von Neumann Morgensten) of course changes with probability and is linear in the corresponding probabilities. The new EU= 0.6u(100)+0.4u(200)=(0.6)(3)+(0.4)(5)=1.8+2=3.8 Notice that the formula for EU=pu(w1)+(1-p)u(w2) is linear in probability, p. Hope this helps.

[@arkarupbasumallik]

Also, I forgot to add that u(100) does not double from 3 to 6 when the probability of getting 100 as wealth doubles. Only p changes from 0.3 to 0.6 and u(100) remains at 3.

[@revelations2044]

I think he meant that it's linear to the expected return. So x is the expected return, then u(x) = constant * x is a linear function of x. Which means that we do not care about risk, only about the expected return

[@yongmrchen]

I think in expected utility theory we assume that the utility (function) of an event or choice does not depend on the probability that that event occurs (it is the utility as if the event is certain), and simply because of this assumption, we can define expected utility as the product of the probability and the utility of the event. This is the same way as we define expected value, in which the value of an event is independent of the probability of the occurrence of the event. If the utility (function) of an event depended on the probability of the occurrence of the event, then the unity function would have two parameters, x and p, i.e., u = u (x, p). It follows that the expected utility would become EU = p*u(x,p). This does not make sense.

[@yongmrchen]

The reason he mentioned linearity in probability is that, I guess, he may discuss decision weight later, which is a function of probability, and that function could be nonlinear.

[@coomassieblue5035]

Horrible educator! I am sure he is a very smart man. But he is barely able to finish a sentence, sometimes. The whole lecture seems like an insecure student who desperately tries to show that he knows much instead of teaching what he knows. Nervously reading slides to the students as fast as one can is a habit of a bad teachers. I hope he watches these videos and tries to become better.

[@montse619XD]

the consistent student participation makes me think he shouldn’t have to break it down as if he were lecturing children… you could slow down the video if you’re unable to keep up lol

[@coomassieblue5035]

@@montse619XD Sure, try to make him look better. But that's the number one excuse of bad teachers: " Smart people would understand" 🙄 I don't expect anything impossible and the subject is not hard. So if he is struggling to sell this simple topic it is his own fault. I really hope that he just had a bad day. But he put this stuff online...so this probably the best that he could do. But it's nice that you are trying to protect him. 👍

[@Stjernberg274]

I've learned a ton from this guy. We should all show some appreciation that these lectures were uploaded and free for all of us to learn from. So far I've only heard people complain about him not speaking with perfect British accent (he's German lol?) and now you his appearance. What about focusing on what he says and what his slides say? To me he's doing a great job at including the class with questions and interactions - a lot better than many lecturers I've had.