# [R] Generating weights based on certain condition in the dataset

Tue Dec 13 17:12:28 CET 2011

```Dear all,

My goal is to create weights.

I have three covariates x1 x2 x3

These three covariates are discrete, they can only store either 1 or 0. So

x1 x2 x3
----------
0  1   0
0  0   1
1  0   0
1  1   0
1  0   1
0  1   1
......

Sometimes they have missing data

x1 x2 x3
----------
0  1   -
0  0   1
1  -   0
1  1   0
-  0   1
0  -   -
......

I need help to generate the following situation
a) If there is a missing data then it can either be 0 0r 1. So lets consider
the first row. The new program should create 2 new rows because x13 is
missing. The two new rows should be like this

x1 x2 x3  weight
----------------
0  1   0      w1=0.5
0  1   0      w2=0.5
0  0   1
1  -   0
1  1   0
-  0   1
0  -   -
......

Consider the last row. The last row contains two missing data. So the new
program should create 4 new rows. The new rows should be like this.

x1 x2 x3  weight
----------------
0  1   0      w1=0.5
0  1   0      w2=0.5
0  0   1
1  -   0
1  1   0
-  0   1
0  0   0     w1 =0.25
0  1   1     w2 =0.25
0  0   1     w3 =0.25
0  1   0     w4 =0.25
......

Also please see the image attached. Please let me know if you have more
questions.

Sincerely