[R-sig-Geo] power index

az14 abidi.zineb at gmail.com
Mon Aug 4 14:31:54 CEST 2014


Dear list, 
 
I'm trying to calculate  Banzhaf power index for a set of 1260
municipalities. 
 My data is as below, where the insee variable is the municipalities, the
variable "nbrseat" is the number of seat in the CM of each municipalities
and the "numeroc" is the groupes ( numeroc takes the value of 1 for the
group 1 ans 2 for the group 2....group 119) 
The program i used is :
data<-read.csv("bzh.csv",header=TRUE,sep=";",dec=',')
attach(data)
data<-as.matrix(data)
summary(data)
for (numeroc in 1 :119) {nbseat<-Delegreel  
                         #Banzhaf power index  
                         #Length of the group : 2^m 
                         subsets<-length(numeroc)-1
                         subs<-matrix(FALSE,2^subsets,subsets)
                         for (i in 1:subsets) {
                               subs[,i] <- rep(c(rep(FALSE, 2 ^ (i - 1)),
                                                   +                      
rep(TRUE, 2 ^ (i - 1))),
                                                 +                     2 ^
(subsets - i))
                            }
                         q<-sum(nbseat)/2
                         banzhaf <- function(i) {
                           other <- nbseat[-i]
                           part.sum <- apply(subs, 1, function(x)
{sum(other[x]) } )
                           sum((part.sum < q) & ((part.sum + nbseat[i]) >q))
                         }}
BZH <- prop.table(sapply(1:length(Delegreel), banzhaf))

but it did not work. 
i will be very grateful for any help. 
Thank you. 
Sincerely. 

Zineb 


data:
insee	nbrseat	numeroc
22001	5	1
22300	3	1
22384	8	1
22047	6	2
22074	4	2
22244	2	2
22313	2	2
22316	2	2
22033	2	3
22158	9	3
22285	2	3
22295	2	3
22298	3	3
29007	4	4
29012	1	4
29081	8	4
29129	1	4
29275	4	4
22097	3	5
22104	2	5
22180	8	5
22315	1	5
22317	1	5
22318	1	5
22342	2	5
22078	1	6
22101	5	6
22247	6	6
22253	1	6
22264	5	6
22383	1	6
29013	1	7
29016	5	7
29018	2	7
29054	3	7
29139	3	7
29141	1	7
29211	3	7
29261	1	7
56009	4	8
56114	3	8
56152	11	8
56241	4	8
35098	5	9
35124	10	9
35249	4	9
35316	3	9
56056	1	10
56080	6	10
56129	7	10
56134	4	10
56227	2	10
56257	3	10
56025	1	11
56043	4	11
56127	12	11
56145	5	11
56208	1	11
56225	1	11
56256	1	11
22005	4	12
22037	1	12
22072	1	12
22129	1	12
22135	9	12
22216	3	12
22354	4	12
22018	1	13
22204	5	13
22212	4	13
22250	4	13
22256	4	13
22269	1	13
22283	3	13
22032	10	14
22036	1	14
22069	1	14
22071	2	14
22239	4	14
22240	4	14
22305	2	14
22312	1	14
22013	10	15
22040	2	15
22088	1	15
22092	1	15
22139	1	15
22156	2	15
22189	2	15
22249	2	15
22271	1	15
22335	1	15
22035	4	16
22056	6	16
22208	5	16
22263	1	16
22274	1	16
22306	2	16
22308	2	16
22352	1	16
22006	1	17
22030	3	17
22034	5	17
22041	1	17
22245	4	17
22254	4	17
22257	1	17
22340	4	17
22023	2	18
22024	1	18
22025	9	18
22031	3	18
22052	1	18
22132	1	18
22138	1	18
22231	1	18
22243	2	18
22320	1	18
22328	1	18
22046	3	19
22066	4	19
22102	2	19
22191	7	19
22292	1	19
22297	2	19
22303	3	19
56046	11	20
56116	6	20
56233	5	20
29033	2	21
29062	4	21
29115	1	21
29123	3	21
29142	1	21
29162	11	21
29053	5	22
29240	3	22
29263	3	22
29302	11	22
22062	2	23
22083	2	23
22122	2	23
22133	2	23
22147	9	23
22148	1	23
22309	1	23
22333	2	23
22371	3	23
35035	1	24
35046	1	24
35048	1	24
35057	3	24
35084	2	24
35160	1	24
35168	10	24
35175	3	24
35311	1	24
22119	1	25
22131	3	25
22207	7	25
22217	1	25
22227	2	25
22228	4	25
22359	1	25
22387	3	25
22009	1	26
22080	2	26
22171	11	26
22203	9	26
22051	1	27
22084	5	27
22175	3	27
22185	7	27
22193	4	27
22341	2	27
56034	11	28
56168	6	28
56258	5	28
22103	2	29
22190	10	29
22213	9	29
22368	1	29
35034	3	30
35044	1	30
35222	6	30
35247	3	30
35248	1	30
35259	3	30
35270	1	30
35291	1	30
35329	2	30
35339	1	30
35354	1	30
35008	2	31
35102	1	31
35125	12	31
35198	2	31
35199	1	31
35200	2	31
35325	1	31
35359	3	31
22085	1	32
22111	1	32
22127	4	32
22195	7	32
22196	3	32
22199	3	32
22347	3	32
56126	5	33
56147	10	33
56195	2	33
56212	5	33

-- 




--
View this message in context: http://r-sig-geo.2731867.n2.nabble.com/power-index-tp7586862.html
Sent from the R-sig-geo mailing list archive at Nabble.com.



More information about the R-sig-Geo mailing list