Observations about prior check issues

Unless the check clearing process were instantaneous, it seems unlikely that the clearings on a given day would be a function of the issues on that same day. 

As a result of this analysis, it seemed worthwhile to include the amounts of checks issued three and four days previous to the day for which the clearings were to be predicted. Just to be on the safe side, days two and five were also included. This resulted in four more terms to be added to the model  

---