how to proxy a variable? (geometrically declining weighted moving average)


I would be very grateful if someone could help me with a problem. Im replicating a study with new panel data and must now proxy a variable by a given equation (from the study):

π_(L,t) =(1-δ)*π_(L,t-1)+ δ*π_(t-1) :underscore means subscript

It says that π_(L,t) is found from a weight on its own lagged value,(1-δ)*π_(L,t-1), and a weight on π_(t-1). The problem is that the only variable I have is the last one, π_(t-1). The dependent variable is unknown, as well as its laggs and the weight, δ. The study says that its the weights, δ, that is to be estimated.

Does anyone know how to solve this? I would be very glad if there is anyone out there with a solution. Using EViews 5 as software program if that is any help!?

Best regards,
Last edited: