Negative values are deleted in log-transforming. How can negative values be log-transformed without losing? Is it wise to make them all positive by adding equal positive numbers to the entire observations before log-transformation? I have learned from answers on my last question about log transformation of ratio variable that it is not a good idea to add value into original values. However, a log transformation of negative values has a different issue (missing).

Specifically, I want to log-transform x in the below in order to address the potential problem of outliers. In this case, in my field, log(x+6 [the smallest negative number]) is a typical choice. Do you agree with this? Or, do you have any other suggestion? I provide detailed information on variable x as follows.

Code:
sum x, det
x
-------------------------------------------------------------
Percentiles Smallest
1% -5 -6
5% -1 -6
10% -1 -6 Obs 712
25% 0 -6 Sum of Wgt. 712

50% 0 Mean .2373596
Largest Std. Dev. 1.21111
75% 1 4
90% 1 4 Variance 1.466788
95% 2 5 Skewness -.8423147
99% 4 5 Kurtosis 10.86292

Code:
graph box x
Array

Code:
dataex x
----------------------- copy starting from the next line -----------------------
Code:
* Example generated by -dataex-. To install: ssc install dataex
clear
input float x
 1
 0
 1
 0
 0
 0
 0
-1
-1
 0
 0
 0
-1
-1
-3
 0
 0
 0
 0
 1
 1
 0
 1
 1
 0
-2
-2
-2
 0
 0
 0
 0
 0
 1
 1
 1
 1
 0
 0
-1
 0
 0
 0
 0
 0
 0
 0
 1
 0
 1
 0
 0
 0
 0
 0
 0
 0
 0
 0
 0
 1
 0
 0
 0
 0
 0
 0
 0
 0
 1
 1
 0
 0
 0
 0
 0
 0
 1
 1
 2
 2
 0
 1
 2
 0
 0
 0
 0
 0
 0
 0
 1
 0
 0
 0
 0
 0
-3
-3
-1
end
------------------ copy up to and including the previous line ------------------

Listed 100 out of 712 observations