Page 2 of 2
Re: Missing Values in the Signal Eq of the State Space Model
Posted: Thu Jun 25, 2009 5:09 pm
by dhanssen
Thank you -- how could you tell this? I do not see any missing values. I'm sorry I'm so slow here.
Re: Missing Values in the Signal Eq of the State Space Model
Posted: Thu Jun 25, 2009 5:10 pm
by EViews Gareth
Log_b has 3 NAs, and log_c has 12.
Just type = @obs(log_b), and you'll see that the number of observations in log_b is not equal to the number of observations in the workfile
Re: Missing Values in the Signal Eq of the State Space Model
Posted: Thu Jun 25, 2009 5:14 pm
by dhanssen
I see!! Thank you!!!
Re: Missing Values in the Signal Eq of the State Space Model
Posted: Fri Jan 08, 2010 3:23 am
by miriamebawy
I already have only 36 observations in my workfile which my professor thinks is too little so i can't disregard missing observations.
My model is:
Y(demand)= B1 + B2 *media Advertising + B3* price of the product + B4 * price of the competitor's price + B5* market share + Error term
All the variables are there for 3 years,except the figures for the media advertising . In many months, the figures were missing..not beacuse for the fact of being missing, but simply because the firm did not spend on advertising in these months...i.e. they were off air (no advertising) (spending was $ 0)
how can i solve this problem without disregarding the missing observations?
Can i specify a certain value for the missing values in advertising like 1 for e.g.?