Trajectories through temporal networks

Applied Network Science

Table 6 Sequential patterns in mobile money use

Use case	Transaction type motif	Number (%)	Value (%)	Duration (h)
Storage/savings	cash-dep \(\rightarrow\) cash-wtd	5.9	23.1	21
Digital transfer	cash-dep \(\rightarrow\) p2p \(\rightarrow\) cash-wtd	5.9	23.5	24
Circ. digital transfer	cash-dep \(\rightarrow\) [p2p] \(\rightarrow\) p2p \(\rightarrow\) cash-wtd	3.6	8.8	67
Cash payment	cash-dep \(\rightarrow\) cash-pay	4.0	12.5	1.5
Circ. cash payment	cash-dep \(\rightarrow\) [p2p] \(\rightarrow\) cash-pay	2.5	4.2	59
Bill payment	cash-dep \(\rightarrow\) bill-pay	4.0	8.8	0.1
Circ. bill payment	cash-dep \(\rightarrow\) [p2p] \(\rightarrow\) bill-pay	3.3	4.2	77
Airtime purchase	cash-dep \(\rightarrow\) mins-pay	40.3	2.4	165
Circ. airtime purchase	cash-dep \(\rightarrow\) [p2p] \(\rightarrow\) mins-pay	21.2	1.2	276

Trajectories observed when tracing funds using the last-in-first-out heuristic, grouped by their sequential transaction patterns. The number and value of trajectories is reported as a percentage of the number and value of all trajectories that begin with a deposit made in the first 5 months of data collection. The median duration is weighted by value. Brackets denote that consecutive person-to-person transactions have been consolidated. Not shown are sequential transaction patterns reflecting niche actions (i.e., below 1% of trajectories) or incomplete trajectories (i.e., money that remains in the system at the end of the finite data collection window). For this reason, the percentages do not add up to 100%