More lags
It might be interesting to learn if coupon actions only cause an immediate effect on sales or if the effect continues into the next period.
Again, you shift the COUPON predictor back by using the function lag(). The result is assigned to a new variable Coupon.lag. Both, the original COUPON and the Coupon.lag variables, are added to the log(SALES) ~ PRICE + Price.lag relationship by using the update() function.
Este exercício faz parte do curso
Building Response Models in R
Instruções do exercício
- Create a lagged variable for
COUPONnamedCoupon.lag. - Update the
lag.modelobject forCOUPONandCoupon.lagby using the functionupdate().
Exercício interativo prático
Experimente este exercício completando este código de exemplo.
# Take the lag of COUPON
Coupon.lag <- ___
# Update the model for COUPON and Coupon.lag
___(lag.model, . ~ . + ___ + ___)