I am looking at the crime data for Nepal and come across this puzzilng point. 

The data for 1993 - 2006 can be found at - here and the data for 2000 to 2009/10 can be found here  .

Crimes are classified according to different types. So I have learnt that the Maoist rebellion caused deaths of about 14,000 people over the 10-year long Civil War from 1996 to 2006. However, adding up the number of murders in this dataset, it doesn't even come close to this number of 14,000. Is this data (political murders during the rebellion) excluded from the list ? Or, is it because a lot of these data (murders by Maoist, Police and Army) were not reported to the police? However, looking at the cases filed at the Supreme Court in the same data files, I find murders to be in the range of 5000 per year. What explains the discrepancy - the data collection method? I would be thankful if anyone could shed light on these issues!