Example Does the shape of the utility function matter? A group of towns intends to develop a resort. Two possible sites have been identified. The council is currently conducting studies on the merits of the two sites. The final decision will be made next year. Land values around the two possible sites are increasing as investors speculate that property values will rise sharply in the vicinity of the new 20.1 Monday 4 July 2022 04:02 AM
20
Embed
Example Does the shape of the utility function matter? A group of towns intends to develop a resort. Two possible sites have been identified. The council.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Example
Does the shape of the utility function matter?
A group of towns intends to develop a resort. Two possible sites have been identified. The council is currently conducting studies on the merits of the two sites. The final decision will be made next year. Land values around the two possible sites are increasing as investors speculate that property values will rise sharply in the vicinity of the new resort.
20.1
Friday 21 April 2023 07:17 AM
Example
A private real estate developer would like to build houses near the new resort. They have identified parcels of land near the two sites. Current prices (£,000) and estimated future values of developable land are summarised below.
20.2
Parcel of land near location
A BCurrent purchase price £1,200 £700Present value if future cash flows if resort is built at this location
£1,500 £1,100
Present value of future sales price of parcel of land if resort is not built at this location
£950 £350
Example
The developers can purchase land now either at site A only, at site B only, or at both. If land is purchased and the resort is not built at that location, the land will sold at a loss. If the land is not purchased now, it will not be available at a later date.
It is assessed that there is a probability of 0.4 that the new resort is built at location A (0.6 at location B).
20.3
Financial Decision Tree
20.4
0.4
Site A s elected Payoffs
£300
Buy A £1,500 ##
-£1,200 -£30 0.6
Site B s elected
-£250
£950 ##
0.4
Site A s elected
-£350
Buy B £350 ##
-£700 £100 0.6
Site B s elected
£400
2 £1,100 ##
£100
0.4
Site A s elected
-£50
Buy A and B £1,850 ##
-£1,900 £70 0.6
Site B s elected
£150
£2,050 ##
Buy nothing
£0
£0 £0
Generating cash flows from left to right
Probabilistic Decision Tree
20.5
0.4
Site A s elected Payoffs
£300
Buy A £1,500 ##
-£1,200 -£30 0.6
Site B s elected
-£250
£950 ##
0.4
Site A s elected
-£350
Buy B £350 ##
-£700 £100 0.6
Site B s elected
£400
2 £1,100 ##
£100
0.4
Site A s elected
-£50
Buy A and B £1,850 ##
-£1,900 £70 0.6
Site B s elected
£150
£2,050 ##
Buy nothing
£0
£0 £0
Generating cash flows from right to left
Combined Decision Tree
20.6
0.4
Site A s elected Payoffs
£300
Buy A £1,500 ##
-£1,200 -£30 0.6
Site B s elected
-£250
£950 ##
0.4
Site A s elected
-£350
Buy B £350 ##
-£700 £100 0.6
Site B s elected
£400
2 £1,100 ##
£100
0.4
Site A s elected
-£50
Buy A and B £1,850 ##
-£1,900 £70 0.6
Site B s elected
£150
£2,050 ##
Buy nothing
£0
£0 £0
So the selected strategy is option 2, to buy B.
Assessment of the Managers Utility
20.7
But does the “shape” of the utility function matter?
Assessment of the Managers Utility
20.8
Enter the value that makes you indifferent to the two values below.
0.7
W ith certain ty
£350 £3,500
£2,240
0.3
-£700
You would accept £350, rather than take the gamble.
Experimentally Observed Utility Function – Low Stimuli
20.13
Using the previous values and the low stimuli Booij and van de Kuillen 2009 model.Where low stimuli reflect the financial levels of the decision the participants had to make.
Decision Tree Employing Regression For Experimentally Observed Utility Function
– Low Stimuli
20.14
0.4Site A selected Payoffs Utilities
£300 0.03Buy A £1,500 0.03
-£1,200 -0.1674 0.6Site B selected
-£250 -0.30£950 -0.30
0.4Site A selected
-£350 -0.45Buy B £350 -0.45
-£700 -0.1436 0.6Site B selected
£400 0.064 £1,100 0.06
0.000.4
Site A selected-£50 0.00
Buy A and B £1,850 0.00
-£1,900 -0.0098 0.6Site B selected
£150 -0.02£2,050 -0.02
Buy nothing£0 0.00
£0 0.0000
Uses linear regression
Low-Stimuli For Data From Booij And Van De Kuilen 2009 Table 5.2
So the selected strategy (just) is option 4, to buy nothing.
Decision Tree Employing Piece Wise Interpolation For Experimentally
Observed Utility Function – Low Stimuli
20.15
0.4Site A selected Payoffs Utilities
£300 0.05Buy A £1,500 0.05
-£1,200 -0.1274 0.6Site B selected
-£250 -0.25£950 -0.25
0.4Site A selected
-£350 -0.42Buy B £350 -0.42
-£700 -0.1250 0.6Site B selected
£400 0.074 £1,100 0.07
0.000.4
Site A selected-£50 -0.04
Buy A and B £1,850 -0.04
-£1,900 -0.0005 0.6Site B selected
£150 0.03£2,050 0.03
Buy nothing£0 0.00
£0 0.0000
Uses piece wise linear interpolation
Low-Stimuli For Data From Booij And Van De Kuilen 2009 Table 5.2
So the selected strategy (just) is option 4, to buy nothing.
Experimentally Observed Utility Function – High Stimuli
U intercept gradient R2positive -0.0261 0.000297 99.52%negative 0.0708 0.001489 98.48%
Select high or low stimuli
Data from Booij and van de Kuilen 2009 Table 5.2
-1.00
-0.80
-0.60
-0.40
-0.20
0.00
0.20
0.40
0.60
0.80
1.00
-1000 0 1000 2000 3000 4000U
x
Data fromBooij and vande Kuilen 2009Table 5.2Fit
Using the previous values and the high stimuli Booij and van de Kuillen 2009 model.Where high stimuli reflect the financial levels of the decision the participants had to make.
Decision Tree Employing Regression For Experimentally Observed Utility Function
– High Stimuli
20.17
0.4Site A selected Payoffs Utilities
£300 0.06Buy A £1,500 0.06
-£1,200 -0.1557 0.6Site B selected
-£250 -0.30£950 -0.30
0.4Site A selected
-£350 -0.45Buy B £350 -0.45
-£700 -0.1246 0.6Site B selected
£400 0.093 £1,100 0.09
0.010.4
Site A selected-£50 0.00
Buy A and B £1,850 0.00
-£1,900 0.0096 0.6Site B selected
£150 0.02£2,050 0.02
Buy nothing£0 0.00
£0 0.0000
Uses linear regression
High-Stimuli For Data From Booij And Van De Kuilen 2009 Table 5.2
So the selected strategy (just) is option 3, to buy A and B.
Decision Tree Employing Piece Wise Interpolation For Experimentally
Observed Utility Function – High Stimuli
20.18
0.4Site A selected Payoffs Utilities
£300 0.06Buy A £1,500 0.06
-£1,200 -0.1257 0.6Site B selected
-£250 -0.25£950 -0.25
0.4Site A selected
-£350 -0.43Buy B £350 -0.43
-£700 -0.1207 0.6Site B selected
£400 0.083 £1,100 0.08
0.000.4
Site A selected-£50 -0.04
Buy A and B £1,850 -0.04
-£1,900 0.0024 0.6Site B selected
£150 0.03£2,050 0.03
Buy nothing£0 0.00
£0 0.0000
Uses piece wise linear interpolation
High-Stimuli For Data From Booij And Van De Kuilen 2009 Table 5.2
So the selected strategy (just) is option 3, to buy A and B.
Experimentally Observed Utility Function – High Stimuli versus Low
Stimuli
20.19
But they hardly differ!
Assessment of the Managers Utility
20.20
But does the “shape” of the utility function matter?