Resumen de Modelling motor temporary disability data with periodics peaks
Lluís Bermúdez, Dimitris Karlis, Miguel Santolino
Studies analyzing temporary consequences after a motor crash are scarcer than those analyzing permanent injuries or mortality. A regression model to evaluate risk factors affecting the duration of temporary disability after motor injury occurs is constructed using a motor insurance dataset. The length, measured in days, of medical leave after a motor accident is used here as a measure of temporary disability severity. The probability function of the number of days with medical sick leave shows spikes in multiplicities of five (working weeks), seven (calendar weeks), thirty (months), etc. To account for such a characteristic of the data, a regression model based on finite mixtures of multiple discrete distributions is proposed to fit the data properly. The model provides a very good fit when working week, week, fortnight and month multiplicities are taken into account. Characteristics of the accident such as gender, age, victim's position and severity of the permanent injuries were found significant to explain the length of temporary disability.
