EViews.com

Hi all, I got trouble comparing two series with unequal observations. My real data is huge so I just create a trial data to work on that at this moment. As in the picture that I attached, originally I have 5 series Analyst, analyst_1, day, day_1 and value_1.
I want to create a new series called Value that takes value of Value_1 if analyst=analyst_1 and day < day_1. The trouble is the comparison is not row by row. For example, if the first line analyst(1) = analyst_1(1) and day(1) > day_1(1), the program will keep compare the next pair day(1) with day_1(2). If the condition is satisfied ( day(1) < day_1(2)), value(1) will equal value_1(2) or 1.1 as in the pic. So for every pair analyst and analyst_1 equal, each obs in day will be compared to each observation in day_1 correspondingly.

This is the code that I write for this trial data

SERIES value

FOR !i=1 TO @obsrange
FOR !j=1 TO @obsrange
IF analyst(!i)=analyst_1(!j) AND day(!i) <= day_1(!j) THEN value(!i)=value_1(!j)

ENDIF
NEXT
NEXT

It works but if I use it for my real data ( which the no of obs is 120 000), the program takes forever to run. I am just wondering if there is a more elegant way to do this.

I am really appreciated if anyone could help.

Hi, thx Gareth
I tried using @recode before but it seems that the comparison is made row by row ( i.e day(1) with day_1(1), day(2) with day_1(2)) rather than between day(1) with each obs of day_1 for the same analyst). I attached the output that I got, the new series value is different from what it is supposed to be as in my previous post.

I'm lost then. Perhaps you could describe the comparison you're trying to make again.