【耶魯大學】《金融市場》（全集）頂尖投資大師課！ P36 - 耶魯大學《金融市場_Financial Markets》中英字幕 - 36.35_粉筆講解：更多關於前景理論.zh_en(Av1450436343,P36)

Professor, I was wondering if we could go over some of the key features of prospect theory at a very high level and just kind of talk about what features it has as well as how it might be different from more traditional approaches. Well, prospect theory is written by two psychologists, not economists, Daniel Kahneman and Amos Tversky over 30 years ago, overturning a key assumption of much economics that represents people as rational and calculating in a sensible way. So what Kahneman and Tversky said is that people, maybe that's a little bit rational, but it looks a little screwy what people do actually. And what they did is a number of experiments that show about decisions under uncertainty that show funny behavior of individuals. So how to present it? The thing that it takes off from is the economist's idea that people are maximizers. We're all calculating. We're all out there figuring out. Now, to some extent, we are. But what is it that we maximize? So economists say utility. Then they also have to recognize that utility is happiness or something like that. But they also recognize that you don't know the outcome before you make your decision. So it's random. So we have to say expected utility. This is traditional economist. I know what I want, and I'm very consistent in pursuing that under uncertainty. But Kahneman and Tversky said, Well, you know, people don't always do that. Let's talk about flipping a coin. I'm going to propose to you. Let's toss a coin. If it comes up heads, I will pay you $200. If it comes up tails, you pay me $100. Want to do it? Yeah, I'm a little hesitant. But it's a positive expected value because it's equally likely the two outcomes. One of them, I pay you $200. But you only pay me $100 if it goes the other way. So you have an expected gain of $50. So you said you're a little hesitant. Since I'm a grad student. You're not going to say no because you know what the right answer, the rational answer is. The rational answer from expected utility theory is of course I'll take it because it has positive expected value. I will take lots of those bets. And they'll all average out to a gain. And so you should take it unless there's something very unusual about your utility. So what Kahneman and Tversky said is that people have a value function. So I'll draw a curve here. It won't be a curve. This is gains on this axis. And this is, they call it value. But it's happiness or utility. And the curve is kinked like this. All right. This is where we are now. The reference point. This is for you. Plus 200. And this is for you. Minus. This is minus 100. All right. So you're looking at your happiness. You're thinking, well, plus 200 only gives me a little happiness. This amount. But losing 100 gives me a lot of unhappiness. So you won't do it. This kink here represents. This is where we are now. In prospect theory, there's always a kink where you are now. You could make $10,000 more in your work, let's say. And it puts you way over here. The kink moves with you. It's always there. So at any point of time, you're always worrying about losses. OK. Relative to where you are now. Yes, relative to where you are now. But this is irrational. Because you should, anytime anyone offers this bet to you, you should always take it. And if you do it many times, you'll come out ahead on average. But people don't. That represents an irrational behavior that Kahneman and Tversky called loss aversion. We worry so much about losing money. We don't have the idea that on average I'll do fine. I just don't. But you don't feel that way. Any time that you make a loss, you kick yourself and you feel bad. So you don't want to take that chance. I see. And that's an irrational behavior. And it's demonstrated that people behave this way. And so, just to reiterate, it's this steep curve that is kind of what is the key to all this. Right. That piece that follows you around. Follows you. You're always worried about little losses today.