Risk preferences

Evidence from a MPL experiment

Author

Me

Published

May 8, 2023

Load data

  • We load the data from the MPL experiment
    • Assign the tibble to d
d <- read_csv("all_apps_wide.csv") |>
    filter(participant._index_in_pages==6) #only those who finished
  • Select only the “relevant” columns
    • Assign it to d.sel
d.sel <- 
d |>
    select(participant.code, session.code, paste("prolific_MPL.1.player.HL_", 1:10, sep = ""))

Session descriptives

  • Number of participants
d.sel |>
distinct(participant.code) |>
    nrow() |>
    kable(caption="Number of participants", col.names = NULL) |>
    kable_styling()
Number of participants
12
  • Number of sessions
d.sel |>
    distinct(session.code) |>
    nrow() |>
    kable(caption = "Number of sessions", col.names = NULL) |>
    kable_styling()
Number of sessions
2

Choices

Frequency of A choices by prospect

  • We compute the frequency of choices by prospect
dt <-
d.sel |>
    select(participant.code, prolific_MPL.1.player.HL_1:prolific_MPL.1.player.HL_10) |>
    pivot_longer(names_to = "Prospect", values_to = "Choice", 2:11) |>
    mutate(Prospect=substr(Prospect,26,27)) |>
    mutate(Prospect=as.integer(Prospect)) |>
    mutate(Choice = ifelse(Choice == 1, "A", "B")) |>
    group_by(Prospect) |>
    count(Choice) |>
    mutate(Frequency = round(prop.table(n),3))
dt |> 
    select(Prospect,Choice, Frequency) |>
    pivot_wider(names_from = Choice, values_from = Frequency) |>
        kable(caption = "Relative frequency of choices by prospect") |>
        kable_styling(font_size = 12, full_width = FALSE, bootstrap_options = c("striped", "hover", "condensed")) 
Relative frequency of choices by prospect
Prospect A B
1 0.750 0.250
2 0.917 0.083
3 0.833 0.167
4 0.583 0.417
5 0.333 0.667
6 0.500 0.500
7 0.250 0.750
8 0.167 0.833
9 0.083 0.917
10 0.167 0.833
dt |>
ggplot(aes(x=Prospect, y=Frequency, fill=Choice)) +
    geom_col() +
    scale_fill_manual(values=c("grey", "black")) +
    labs(x="Prospect", y="Frequency", fill="Choice") +
    scale_x_continuous(breaks = c(1,2,3,4,5,6,7,8,9,10)) +
    theme_bw() +
        labs(
            title = "Frequency of choices by prospect",
            y = "Relative frequency",
            x = "Prospect",
            caption = "Source: MPL experiment",
            color = "Odd"
        ) +
        theme(
            legend.position = "bottom",
            axis.text = element_text(size = 8),
            axis.title = element_text(size = 14, face = "bold"),
            legend.background = element_rect(
                fill = "grey",
                linewidth = 2, linetype = "solid"
            )
        )