Simulation experimental design Firm Choice

Published

March 24, 2026

The following code shows the selected priors and the corresponding WTP values as well as the deterministic choices

The simulation has 300 respondents and 2000 runs. The simulation itself took 26M 3.58100000000013S .

Prior selection and deterministic utility.

The following priors have been selected and they result in the following WTP values

# Priors chosen in this simulation
b_flood <- all_designs$arguements$`Beta values`$bflood
b_heat <- all_designs$arguements$`Beta values`$bheat
b_tax <- all_designs$arguements$`Beta values`$btax


(c(b_flood,b_heat,b_tax))

[1]  0.200  0.150 -0.016

# Calculate WTP values
wtp_flood <- -b_flood / b_tax
wtp_heat <- -b_heat / b_tax


# Print the WTP values
cat("\n ________________ \n WTP for 10 % (one unit) Flood Reduction: ", wtp_flood, "\n 50 % Increase: ", 5*wtp_flood,  "\n 75 % Increase", 7.5*wtp_flood)


 ________________ 
 WTP for 10 % (one unit) Flood Reduction:  12.5 
 50 % Increase:  62.5 
 75 % Increase 93.75

cat("\n ________________ \n WTP for 10 % (one unit) Heat Reduction: ", wtp_heat, "\n 30 % Increase: ", 3*wtp_heat,  "\n 50 % Increase", 5*wtp_heat)


 ________________ 
 WTP for 10 % (one unit) Heat Reduction:  9.375 
 30 % Increase:  28.125 
 50 % Increase 46.875

For a 1% reduction in flood risk respondents are willing to accept a tax increase 12.5 percentage points. 50% reduction in flood risk is worth a 62.5 percentage point increase in taxes.

Statistics and power

Here you see the statistics of the parameters for 2000 runs.

kable(all_designs[["summaryall"]] ,digits = 3) %>% kable_styling()

parname	utilitybayesian.n	utilityfixed.n	truepar	utilitybayesian.mean	utilityfixed.mean	utilitybayesian.sd	utilityfixed.sd	utilitybayesian.min	utilityfixed.min	utilitybayesian.max	utilityfixed.max	utilitybayesian.range	utilityfixed.range	utilitybayesian.se	utilityfixed.se	utilitybayesian.median	utilitybayesian.skew	utilitybayesian.kurtosis	utilityfixed.median	utilityfixed.skew	utilityfixed.kurtosis
bflood	2000	2000	0.200	0.201	0.200	0.026	0.026	0.111	0.097	0.295	0.307	0.184	0.210	0.001	0.001	0.200	0.031	-0.009	0.200	0.077	0.151
bheat	2000	2000	0.150	0.151	0.151	0.032	0.032	0.052	0.043	0.271	0.257	0.220	0.214	0.001	0.001	0.151	0.006	0.005	0.151	0.041	-0.041
btax	2000	2000	-0.016	-0.016	-0.016	0.008	0.007	-0.041	-0.040	0.008	0.007	0.050	0.047	0.000	0.000	-0.016	-0.066	-0.091	-0.016	-0.089	-0.070
boptout	2000	2000	1.000	1.007	1.002	0.180	0.180	0.225	0.340	1.601	1.524	1.376	1.184	0.004	0.004	1.007	0.016	-0.013	0.997	0.128	0.012
rob_pval0_bflood	2000	2000	NA	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	NaN	NaN	0.000	NaN	NaN
rob_pval0_bheat	2000	2000	NA	0.001	0.001	0.005	0.005	0.000	0.000	0.100	0.150	0.100	0.150	0.000	0.000	0.000	13.430	210.105	0.000	18.987	450.278
rob_pval0_btax	2000	2000	NA	0.133	0.119	0.207	0.194	0.000	0.000	1.000	0.990	1.000	0.990	0.005	0.004	0.040	2.117	4.031	0.030	2.330	5.220
rob_pval0_boptout	2000	2000	NA	0.000	0.000	0.004	0.001	0.000	0.000	0.190	0.050	0.190	0.050	0.000	0.000	0.000	44.476	1982.005	0.000	36.879	1490.002

all_designs[["powa"]]

$utilitybayesian

FALSE  TRUE 
46.45 53.55 

$utilityfixed

FALSE  TRUE 
 45.9  54.1

Illustration of simulated parameter values

To facilitate interpretation and judgement of the different designs, we plot the densities of simulated parameter values from the different experimental designs.

$bflood


$bheat


$btax


$boptout