Discussions

Hi Dewey,

I feel very confused about the visit counts variables in weekly pattern and weekly pattern plus datasets. Could you help clarify it?

1. In Weekly Patterns, RAW_VISIT_COUNTS is the observed raw number of visits. NORMALIZED_VISITS_BY_STATE_SCALING, based on its description, is raw_visit_counts scaled using the mobile device sampling rate for the state in which the POI is located. So NORMALIZED_VISITS_BY_STATE_SCALING is raw_visit_counts × (total population in each state or province/observed total visits within the state or province in that week)? If this is the case, the two variables TOTAL_VISITS and TOTAL_DEVICES_SEEN in Normalization Stats / Weekly Patterns are not needed since NORMALIZED_VISITS_BY_STATE_SCALING is already a state-level normalized variable and more precise than country-level normalization.
2. In Weekly Patterns Plus, there is VISIT_COUNTS but no NORMALIZED_VISITS_BY_STATE_SCALING any more. VISIT_COUNTS is said in https://docs.deweydata.io/docs/weekly-patterns-plus to be normalized as Estimated Visits = ( devices_store / devices ) × country_population. If this is the case, how is VISITOR_COUNTS normalized in Weekly Patterns Plus?

Thanks,

Xin