Abstract
This
paper constructs a panel smooth transition autoregression model to evaluate the
impact of the U. S.
business cycle on the stability of lottery sales. We
find that the impact is nonlinear and time-varying,
depending on the level of the leading indicator. The persistence of lottery sales locates
between 0.7779 and 0.8539, implying that current lottery sales are influenced
by about 15%-23% of current disturbance, and the stability of the lottery sales
is relatively high. An increase in the
leading indicator would lead to a higher persistence of lottery sales and then more
stable lottery sales in the next period.
JEL
classification numbers: C23, L83, E32
Keywords:
Sales
persistence,
stability of lottery sales, panel smooth transition autoregression
(PSTAR) model, lagged leading indicator