1
$\begingroup$

Lottery is between:

Option A: a certain choice of £5

Option B: £10 with probability 0.1 and £1 with probability 0.9

The probability of receiving £10 increases in each subsequent choice.

How do I calculate the CRRA bounds of this, using U(x) = x^(1-r)/(1-r)

$\endgroup$
1
$\begingroup$

Let me rewrite your Option B as follows: £10 with probability $p$ and £1 with probability $1-p$ where $p\in[0,1]$ varies across the items in the list of options.

Let $\bar p$ be the point where the subject's choice switches. For example, subject chooses Option A for all $p<\bar p$ and Option B for all $p>\bar p$. Then we can use the indifference condition \begin{equation} U(5)=\bar pU(10)+(1-\bar p)U(1) \end{equation} to infer the risk aversion parameter $r$.

In experiments where $p$ is discretized, we can take $p_0,p_1$ as the values that are immediately before and after the switching point in choice list. For example, subject chooses Option A for all $p\le p_0$ and Option B for all $p>p_1$. Then the bounds on the risk aversion parameter can be inferred from the following inequalities: \begin{equation} U(5)\ge p_0U(10)+(1-p_0)U(1) \quad\text{and}\quad U(5)\le p_1U(10)+(1-p_1)U(1). \end{equation}

$\endgroup$
  • 1
    $\begingroup$ +1 NB To infer $r$, we also need to know your initial level of wealth. $\endgroup$ – afreelunch Jan 2 at 12:19
  • $\begingroup$ @afreelunch: You're right theoretically. In actual experimental settings, however, since subjects' initial wealth is difficult to observe, I believe experimenters typically assume it to be zero (or assume utility is reference-dependent, where the reference point being zero). $\endgroup$ – Herr K. Jan 2 at 20:06

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.