Muhammad Asali, Aamer Abu-Qarn, and Michael Beenstock, "The Cycle of 
Violence in the Second Intifada: Causality in Nonlinear Vector 
Autoregressive Models", Journal of Applied Econometrics, Vol. 32, No. 6, 
2017, pp. 1197-1205.

The dataset is based on datasets from the Jaeger-Paserman study (American
Economic Review, 2008) and the Haushofer, Biletzki, and Kanwisher study
(PNAS, 2010).

All data are in the file Intifada_extended_data.csv, which is an ASCII
file in DOS format. It is zipped in the file aab-data.zip. Unix/Linux
users should use "unzip -a".

The data set includes 49 variables and 2665 observations. The
variables are:

date:    The actual date (in which the fatalities, if any, are observed)

pal_tot: Total Palestinian fatalities on day "date"

isr_tot: Total Israeli fatalities on day "date"

period:  Political period in Israel:
1.	Barak government 
2.	Sharon government, before Sep/11
3.	Sep/11 to Oslo accords 
4.	Oslo accords
5.	Post-Oslo, before ceasefire 
6.	Ceasefire
7.	Post Ceasefire
8.	Post Jaeger-Paserman sample, 2005-2007

Period(j): Dummy variables for categories in the "period" variable

completed: The completed length of the separation barrier (km)

dayofweek: Day of the week of the observation (on "date")
sunday	                 
monday                
tuesday 
wednesday 
thursday    
friday 

i1-i14: The lagged Israeli fatalities. That is, i1=isr_tot(t-1),
        i2=isr_tot(t-2), ... , i14=isr_tot(t-14)

p1-p14: The lagged Palestinian fatalities 

ISR:   is zero if there iare no Israeli fatalities; is 1 if there is 1
       Israeli fatality; and is 2 if there are 2 or more Israeli fatalities

PAL:   is zero if there are no Palestinian fatalities; is 1 if there is
       1 Palestinian fatality; and is 2 if there are 2 or more Palestinian
       fatalities