Moving average of lagged values in panel data

[Thanos Kostopoulos]

Dear forum users,

I am working on a panel dataset where country is the cross-sectional unit and year the time dimension. Let me call the variable of interest X. My goal is to calculate country specific means of different window sizes. For instance, for the year 2000 in the Netherlands, I would like to calculate the Netherlands specific mean of X during the years 1996 - 2000 for a 5 year window, or the mean of the years 1991 - 2000 for a 10 year window, etc. There are no missing values. Can someone help me with this?

Thank you very much in advance!

ps: I tried to simplify the context but please let me know if more information is needed.

[Phil Bromiley]

Welcome to Stata list. You will increase your chances of useful answer by providing Stata code in code delimiters, readable Stata output, and sample data using dataex.

Knowing the exact structure of your data is essential to helping you really. If you have panel data, then you need to xtset your data. If you xtset your data, you can refer to previous values with lags. So the mean of the previous five values could be
g m=(L.x+ L2.x + L3.x + L4.x+L5.x)/5

There are more efficient ways to program it but this is an easy way to understand for a beginner.

[Attaullah Shah]

You can also try asrol for various rolling window statistics.

ssc install asrol
help asrol

*some examples

*Example 1: Find the arithmetic mean in a rolling window of 4 observations for each group (company)
 
    . webuse grunfeld
 
    bys company: asrol invest, stat(mean) win(year 4)

    *This command calculates the arithmetic mean for the variable invest using a four years rolling window and stores the results in a new variable, mean4_invest.

*Example 2: Geometric mean in a rolling window of 10
 
    . webuse grunfeld
 
    bys company: asrol invest, stat(gmean) win(year 10)