Strange Error in Complex program

[Matuk]

Hello,
I'm currently working on my dissertation paper, which uses some filtering techniques and a Consensus measure to aggregate them, following Darvas and Vadas (2005). I'm trying to use the author's eviews program (which works fine with his data set) but when i try to use my own input from the xls. the program suddenly stops, while doing the Beveridge-Nelson Decomposition (at the estimating the ARMA models) and the status line reads: "Too few observations for line". I've tried it with different data sets but nothing seems to work...
I've attached the program and the .xls with my own data set, and hope that your expertise can find out what the issue is.

Thank you very much for your time,
Matuk

[Matuk]

EDIT:
I've changed my dataset, now the program works for more data but received a new error: "Insufficient number of observations in "EQUATION EQ_QT.LS Q C TREND @TREND*@TREND".Now i really have no idea what is wrong.

Thanks again,
Matuk

[EViews Gareth]

You probably have too few observations for the equation to be run. How many observations does Q and TREND have for the current sample?