To analyse the functional relationship between electroencephalography (EEG) and facial electromyography (EMG), Gentsch et al. (2014) simultaneously recorded EEG and EMG signals from 24 participants while they were playing a computerised gambling task. The given subset contains aggregated observations of 23 participants. Curves were averaged over each subject and each of the 8 study settings, resulting in 23 times 8 curves.

`data("emotion")`

A list with the following 10 variables.

`power`

factor variable with levels

*high*and*low*`game_outcome`

factor variable with levels

*gain*and*loss*`control`

factor variable with levels

*high*and*low*`subject`

factor variable with 23 levels

`EEG`

matrix; EEG signal in wide format

`EMG`

matrix; EMG signal in wide format

`s`

time points for the functional covariate

`t`

time points for the functional response

The aim is to explain potentials in the EMG signal by study settings as well as the EEG signal (see Ruegamer et al., 2018).

# NOT RUN { data("emotion", package = "FDboost") # fit function-on-scalar model with random effect and power effect fos_random_power <- FDboost(EMG ~ 1 + brandomc(subject, df = 2) + bolsc(power, df = 2), timeformula = ~ bbs(t, df = 3), data = emotion) # } # NOT RUN { # fit function-on-function model with intercept and historical EEG effect # where limits specifies the used lag between EMG and EEG signal fof_historical <- FDboost(EMG ~ 1 + bhist(EEG, s = s, time = t, limits = function(s,t) s < t - 3), timeformula = ~ bbs(t, df = 3), data = emotion, control = boost_control(mstop = 200)) # }