EWMA Questions

[willie-shen]

I just wanted to ask, how do I go from

16y(t) = x(t) + 15y(t-1)

to

16y(t) = x(t) + 16y(t-1) - (16(y(t-1)-16/2)/16

My understanding is that if I apply the correction factor for rounding, wouldn't it be

16y(t) = x(t) + 15y(t-1) + 8