We introduce a class of multivariate latent Markov models with covariates for the investigation of criminal trajectories. For the maximum likelihood estimation of these models we outline an EM-type algorithm. We also show how, by fitting a suitable sequence of nested models belonging to the proposed class, we can perform a hierarchical classification of the crimes into homogeneous groups.
Bartolucci, F., Pennoni, F. (2005). A class of multivariate latent Markov models for clustering patterns of criminal activity. In Book of Short Papers Meeting of the classification and data analysis group of the Italian Statistical Society (pp.237-240).
A class of multivariate latent Markov models for clustering patterns of criminal activity
PENNONI, FULVIA
2005
Abstract
We introduce a class of multivariate latent Markov models with covariates for the investigation of criminal trajectories. For the maximum likelihood estimation of these models we outline an EM-type algorithm. We also show how, by fitting a suitable sequence of nested models belonging to the proposed class, we can perform a hierarchical classification of the crimes into homogeneous groups.
| File | Dimensione | Formato | |
|---|---|---|---|
|
ClaDAG_Pennoni_2005_index.pdf
accesso aperto
Dimensione
881.56 kB
Formato
Adobe PDF
|
881.56 kB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
