This thesis uses lifelong labor market histories from the 1979 National Longitudinal Survey of Youth to study how occupational mobility relates to the duration of nonemployment (or jobless) spells. An empirical model which takes into account the duration of all spells experienced by an individual is estimated using Bayesian methods. The main nding is that the timing of the decision to change occupations matters when it comes to jobless spells duration. Those spells which immediately precede an occupation change take longer to exit from. On the other hand, spells beginning after one's rst occupation change have shorter durations. Posterior predictive checks reveal that the model underestimates the dispersion of nonemployment duration which is observed in the data.