Michele is currently working on accounting for spatial correlation in damage models. Observations close by tend to be more related than those further away. Although this can be partially explained by factors like ground motion intensity, residual correlation can persist due to unknown factors and result in the underestimation of damage extremes for regional portfolios. Since hazard measurements are often spatially clustered, Michele is also researching the effects of such sampling schemes on statistical inferences.
Prior to starting at NTU, Michele worked in the Malaria Atlas Project based in the University of Oxford where she developed spatial, temporal and spatiotemporal models to support global as well as country-specific epidemiological decisions.