Event Details

On aggregation of strongly dependent network flows

Statistical analysis of networks is often based on aggregated series where aggregation is defined via routing matrices. Granger (1980) showed that aggregation of short-memory processes can imply long-range dependence. On the other hand, OD-flows (origin destination flows) often exhibit long memory. Thus, routing of OD-flows leads to cross-sectional aggregation of strongly dependent series. Asymptotically, dependence increases substantially, transforming a hyperbolic decay of autocorrelations to a slowly varying rate. This makes statistical inference highly uncertain. The situation changes, when time-dependent aggregation is applied. Suitably chosen time-dependent routing schemes can preserve a hyperbolic rate or even eliminate autocorrelations completely. This is joint work with Haiyan Liu and Sucharita Ghosh.