My data has multiple observations within participants.

Each participant has 10 trials.

For each trial, there are 4 options, there are variables associated with each of the options. The participant select one option at a time for 3 times. Each time one option is selected, it is removed and the participant need to select from the remaining ones. The variables associated with remaining choices will change. So my data looks like below. There is no subject level variables.

Since my data is nested, with repeated observations within individuals, the examples of the simple mlogit models (each person makes only one choice) online may not apply to my data. What is the appropriate way to analyze my data?

Each participant has 10 trials.

For each trial, there are 4 options, there are variables associated with each of the options. The participant select one option at a time for 3 times. Each time one option is selected, it is removed and the participant need to select from the remaining ones. The variables associated with remaining choices will change. So my data looks like below. There is no subject level variables.

Since my data is nested, with repeated observations within individuals, the examples of the simple mlogit models (each person makes only one choice) online may not apply to my data. What is the appropriate way to analyze my data?

Code:

```
ID Trial C_num option var1 var2 choice
S1 1 1 C 0.969311934 -0.328713815 1
S1 1 1 D 0.368459569 -0.100616452 0
S1 1 1 A 0.402664843 -0.194083578 0
S1 1 1 B 0.389439842 -0.257392906 0
S1 1 2 D 0.150329117 -0.097984944 0
S1 1 2 A 0.302840263 0.209900197 1
S1 1 2 B 0.338795387 0.122258074 0
S1 1 3 D 0.216206285 -0.11890087 0
S1 1 3 B 0.297109623 0.250219284 1
S1 2 1 C 0.79171215 -0.178453098 0
S1 2 1 D 0.168236191 -0.099548578 0
S1 2 1 A 0.37572025 -0.027878373 1
S1 2 1 B 0.278136821 0.265868384 0
S1 2 2 D 0.17707435 -0.125174593 0
S1 2 2 C 0.7098889 -0.265653 0
S1 2 2 B 0.349678638 0.209010144 1
S1 2 3 D 0.174723298 -0.099354708 0
S1 2 3 C 0.808463668 -0.136589522 1
```

Last edited: