Thoughts on week 7

We didn't have a class this week as David was away at the annual conference of the Society for Judgment and Decision Making. This took me back to my readings of Nassim Nicholas Taleb's books "The Black Swan" and "Fooled by Randomness" over the summer, in one of which he mentions attending just such conference. I'm wondering if David came across him this time? By the way, I can thoroughly recommend both books.

Our group used the time last week to have a meeting to discuss our group wiki page and what needed to be done to complete it before the extended deadline. I don't know about everyone else, but I've found creating the wiki page an interesting challenge. It's very different from writing essays or other types of group work that I've done so far, but overall I think I like it as a form of assessment. I think it also makes group work a bit easier because so much can be done in your own time and the ability to leave comments is a good replacement for face to face meetings, which can sometimes be difficult just from the point of view of getting people together.

So, it's on to the next wiki page...

Thoughts on week 6

This week we looked at the subject of "preference and choice", specifically, how we judge the monetary value of objects in order to make decisions about the object based on this judgment.

The paper reviewed by our group was "Aspects of endowment: a query theory of value construction", Johnson, Haubel & Keinan (2007). The authors present a memory-based account of endowment, which suggests that the value of an object is decided by the order and type of questions (aspects) about an object that we think about when arriving at a valuation.

In summary, their findings showed:

• Buyers produce more value-decreasing aspects (positive thoughts about the money and negative thoughts about the mug)
• Sellers produce more value-increasing aspects (mug: positive, money: negative)
• Buyers and sellers think about aspects in different orders
• The endowment effect can be eliminated by asking buyers and sellers to think about aspects in an order that is opposite to that which they seem to normally use
• The endowment effect can be produced, even without anyone possessing a mug, just by asking buyers and sellers to think about aspects in specific orders

The other papers presented this week dealt with various endowment effect experiments (Kahneman, Knetsch & Thaler (1990)) and a comparison of the Neoclassical Theory, which seems to be used by experts and Prospect Theory, which seems to be used by people who are inexperienced in a particular area (List (2004)).

Interesting thoughts from all of this:

• An example was given of how different people could consider the value of a bottle of wine, with a wine drinker having very different considerations from an investor in wine. This highlights how two people both possessing the same bottle of wine might reach very different valuations. Previously the comparison has been between different valuations arrived at by buyers and sellers.
• The housing market also provides some interesting examples. Consider when there are a lot of "buy to let" investors in the housing market. They might value a property very differently from someone who wants to buy it to live in it. (Can compare back to the wine drinker and the wine investor). This will effect the housing market. A second example can be seen from a falling house price market, as seen recently. Sellers seemed to be unwilling to accept a drop in price, viewing this as a loss even though they may still be in a total gain situation based on the price that they paid for the property. However, the market seemed to start moving again once people realised that a loss on the property that they are selling becomes a gain on the next property that they buy, which would be at a reduced price.

All very interesting and shows the value of knowing as much as you can about someone else's position, if you're ever in a negotiating situation. Of course, in the real world...

More thoughts on week 4

David, following your further comments on my second graphs, I have now produced two corrected ones (based on the answers I would have given if I had read the questions properly), as shown above.

The previous Certainty Equivalence graph had that odd kink in it because I did not enter the monetary values in order. The previous Probability Equivalence graph only had 4 points because there were two values for 500 at .65 (due to reading the question incorrectly).

Further thoughts on weeks 2 & 3

Returning to the topic of "fast and frugal" decision making, I want to pick up on David's comment.

If I were describing Franklin's Rule and the Matching Heuristic to someone who didn't know anything about the subject matter, I might try something like the following...

...if you think of a "cue" as an individual piece of information that is relevant to a particular decision, then, if you were using Franklin's rule to come to a decision, you would consider every single cue and it's importance (what we call "weighting") and combine all of this information to make the decision. However, if you were using the Matching Heuristic, you would only search through some of the cues and base your decision on the first one that you came to that seemed to have a particularly relevant value...

Comments sought on how useful this description is...

Further thoughts on week 4

Here's two further graphs, which as well as showing utility measurement, also reflect very well a measure of stupidity (mine!).

Following David's comments on my "Thoughts on week 4", I revisited the graph producing exercise and produced these initially rather interesting graphs for comparison. However, on considering why they are so different and the graph for certainty equivalence so odd, it would appear that the certain answer is that I didn't read question 3 under the certainty equivalence measure properly (I answered it as if it read exactly the same as question1).

The utility value of this blog as a learning tool has just increased substantially...

Thoughts on week 5

This week we moved on to the subject of "decision framing".

Our group studied the Kahneman & Tversky (1984) paper, "Choices, Values and Frames". This is a really good paper to start with on the subject. The abstract provides a succinct summary of their findings:

• Risk aversion for gains
• Risk seeking for losses
• Overweighting of sure things and those with very small probabilities
• Invariance: different descriptions of the same outcomes lead to different choices. This shouldn't be the case under the invariance criterion
• Negative outcomes might be more acceptable if described as a cost rather than a loss
• The value of an experience is important

The relevance of these findings can be applied in numerous situations. For example:

• Politics: describing outcomes in the most advantageous way for your particular cause

The presentations from the other groups highlighted the further work that has been done with regard to framing effects and some of the contradictory results.

All worth knowing if you ever need to persuade people to take a particular course of action. Unless, of course, they're psychology students and fully aware of such tactics!

Thoughts on week 4

This week we have been reviewing the subject of "Decision making under risk and uncertainty", looking at utility measures, expected utility theory and psychological theories.

Again, there were a few definitions that I picked up, which I find useful to remember:

• Utility - subjective value (as opposed to objective value), a measure of pleasure/displeasure or of usefulness


• Risk - probabilities that are known


• Uncertainty - probabilities that aren't known


• Prospect Theory - outcomes are coded as gains or losses from a reference point. The reference point is subjective & moveable and might be an expectation or an aspiration. A gain might actually be a psychological loss

We also looked at the Priority Heuristic, which I think can be thought of as another "fast & frugal" way of decision making.

An exercise was provided to measure our own utility function. The graph of my results is above (couldn't find out how to move it!). The measure looks at certainty equivalence and probability equivalence, with both methods producing the same results. Not quite the case for my graph! Considering why my 2 measures are not the same, I recalled that when I answered the first set of questions (certainty equivalence) I felt as if I was putting in the "right" answers, i.e. what was logical. When I answered the second set of questions (probability equivalence), I had to think much more about this and was less certain of my answers. I also felt that I needed to amend my answers to the first set of questions, to answer them more "truthfully" than rationally. A very interesting exercise, which I'll go back and do again sometime to see if I answer differently having been through this current thought process. In the meantime, no more snappy decisions!!