TOP 25 COMPANIES BY GAME REVENUES
Jonathan Bouchet
January 13, 2017
library(ggplot2)
library(dplyr)
library(gridExtra)
library(zoo)
library(RColorBrewer)Introduction
- scrapped data from Newzoo
- a bit of manipulation (regexp, reshape dataframe)
- some plots
Data
df<-read.csv('TopGameCompanies_2012.csv',sep=',',header=TRUE)
colnames(df)<-c('image','rank','Company','Q1_2012','Q2_2012','Q3_2012','Q4_2012','YEAR_2012','CHANGE_2012')
df$CHANGE_2012<-as.numeric(gsub("[[:punct:]]","",as.character(df$CHANGE_2012)))
df0<-read.csv('TopGameCompanies_2013.csv',sep=',',header=TRUE)
colnames(df0)<-c('image','rank','Company','Q1_2013','Q2_2013','Q3_2013','Q4_2013','YEAR_2013','CHANGE_2013')
df0$CHANGE_2013<-as.numeric(gsub("[[:punct:]]","",as.character(df0$CHANGE_2013)))
df1<-read.csv('TopGameCompanies_2014.csv',sep=',',header=TRUE)
colnames(df1)<-c('image','rank','Company','Q1_2014','Q2_2014','Q3_2014','Q4_2014','YEAR_2014','CHANGE_2014')
df1$CHANGE_2014<-as.numeric(gsub("[[:punct:]]","",as.character(df1$CHANGE_2014)))
df2<-read.csv('TopGameCompanies_2015.csv',sep=',',header=TRUE)
colnames(df2)<-c('image','rank','Company','Q1_2015','Q2_2015','Q3_2015','Q4_2015','YEAR_2015','CHANGE_2015')
df2$CHANGE_2015<-as.numeric(gsub("[[:punct:]]","",as.character(df2$CHANGE_2015)))
df3<-read.csv('TopGameCompanies_2016.csv',sep=',',header=TRUE)
colnames(df3)<-c('image','rank','Company','Q1_2016','Q2_2016','Q3_2016','Q4_2016','YEAR_2016','CHANGE_2016')
df3$CHANGE_2016<-as.numeric(gsub("[[:punct:]]","",as.character(df3$CHANGE_2016)))
head(df1)## image rank Company Q1_2014 Q2_2014 Q3_2014
## 1 Tencent.jpg 1 Tencent 1673 1785 1824
## 2 Sony.jpg 2 Sony* 1392 1200 1046
## 3 microsoft.jpg 3 Microsoft* 1228 970 1017
## 4 ea.jpg 4 EA 1123 1214 990
## 5 Activision-Blizzard.jpg 5 Activision Blizzard 1111 970 753
## 6 Apple.jpg 6 Apple* 769 859 897
## Q4_2014 YEAR_2014 CHANGE_2014
## 1 1928 7211 37
## 2 1484 5121 8
## 3 1428 4643 5
## 4 1126 4453 22
## 5 1575 4409 4
## 6 947 3472 46head(df2)## image rank Company Q1_2015 Q2_2015 Q3_2015
## 1 Tencent.jpg 1 Tencent 2053 2000 2210
## 2 microsoft.jpg 2 Microsoft* 1416 1463 1517
## 3 Sony.jpg 3 Sony* 1341 1134 1435
## 4 Activision-Blizzard.jpg 4 Activision Blizzard 1278 1044 990
## 5 Apple.jpg 5 Apple* 993 1096 1150
## 6 ea.jpg 6 EA 1185 1203 815
## Q4_2015 YEAR_2015 CHANGE_2015
## 1 2463 8725 21
## 2 2013 6409 38
## 3 1983 5892 15
## 4 1353 4665 6
## 5 1193 4432 28
## 6 1070 4273 4str(df1)## 'data.frame': 25 obs. of 9 variables:
## $ image : Factor w/ 25 levels "Activision-Blizzard.jpg",..: 23 20 14 7 1 2 9 12 18 24 ...
## $ rank : int 1 2 3 4 5 6 7 8 9 10 ...
## $ Company : Factor w/ 25 levels "Activision Blizzard",..: 23 20 14 7 1 2 9 12 18 24 ...
## $ Q1_2014 : int 1673 1392 1228 1123 1111 769 516 607 361 235 ...
## $ Q2_2014 : int 1785 1200 970 1214 970 859 556 594 333 437 ...
## $ Q3_2014 : int 1824 1046 1017 990 753 897 629 514 430 151 ...
## $ Q4_2014 : int 1928 1484 1428 1126 1575 947 661 546 969 983 ...
## $ YEAR_2014 : int 7211 5121 4643 4453 4409 3472 2362 2260 2092 1806 ...
## $ CHANGE_2014: num 37 8 5 22 4 46 71 20 13 33 ...#merge datasets for years 2014, 2015, and 2016
merge1<-merge(df,df0,by="Company")
merge2<-merge(df1,df2,by="Company")
merge3<-merge(merge1,merge2,by='Company')
mergedData<-merge(df3,merge3,by='Company')
#select EA and Tencent companies
#selectedData<-filter(mergedData,Company=='EA' | Company=='Tencent') %>% #select(Company,Q1_2014,Q2_2014,Q3_2014,Q4_2014,Q1_2015,Q2_2015,Q3_2015,Q4_2015,Q1_2016,Q2_2016,Q3_2016,Q4_2016,YEAR_2014,YEAR_2015,YEAR_2016)
#selectedData<- mergedData %>% select(-c(rank,image,rank.x.x,image.x.x,rank.y.x,image.y.x,rank.x.y,image.x.y,rank.y.y,image.y.y))
selectedData<- mergedData %>% select(-c(CHANGE_2012,CHANGE_2013,CHANGE_2014,CHANGE_2015,CHANGE_2016,YEAR_2012,YEAR_2013,YEAR_2014,YEAR_2015,YEAR_2016,rank,image,rank.x.x,image.x.x,rank.y.x,image.y.x,rank.x.y,image.x.y,rank.y.y,image.y.y,Q3_2016,Q4_2016))
#transpose the dataset and rename column names
selectedDataT<-as.data.frame(t(selectedData))
names <- rownames(selectedDataT)
rownames(selectedDataT) <- NULL
summary <- cbind(names,selectedDataT)
companies<-selectedData$Company
t<-c("date")
t<-append(t,as.character(companies),1)
colnames(summary)<-t
summary<-summary[-c(1),]
#colnames(summary)<-c("date","EA","Tencent")
#summary<-summary[-c(1,14:15),]#define a function to convert the quaterly data as Date format (a bit brute-force)
changeDate<-function(x){
if(x=="Q1_2012") {return(as.Date("2012-02-01"))}
else if (x=="Q2_2012") {return(as.Date("2012-05-01"))}
else if (x=="Q3_2012") {return(as.Date("2012-08-01"))}
else if (x=="Q4_2012") {return(as.Date("2012-11-01"))}
else if (x=="Q1_2013") {return(as.Date("2013-02-01"))}
else if (x=="Q2_2013") {return(as.Date("2013-05-01"))}
else if (x=="Q3_2013") {return(as.Date("2013-08-01"))}
else if (x=="Q4_2013") {return(as.Date("2013-11-01"))}
else if (x=="Q1_2014") {return(as.Date("2014-02-01"))}
else if (x=="Q2_2014") {return(as.Date("2014-05-01"))}
else if (x=="Q3_2014") {return(as.Date("2014-08-01"))}
else if (x=="Q4_2014") {return(as.Date("2014-11-01"))}
else if (x=="Q1_2015") {return(as.Date("2015-02-01"))}
else if (x=="Q2_2015") {return(as.Date("2015-05-01"))}
else if (x=="Q3_2015") {return(as.Date("2015-08-01"))}
else if (x=="Q4_2015") {return(as.Date("2015-11-01"))}
else if (x=="Q1_2016") {return(as.Date("2016-02-01"))}
else if (x=="Q2_2016") {return(as.Date("2016-05-01"))}
else if (x=="Q3_2016") {return(as.Date("2016-08-01"))}
else if (x=="Q4_2016") {return(as.Date("2016-11-01"))}
}
summary$newDate<-sapply(summary$date,changeDate)
summary$newDate<-as.Date(summary$newDate)
summary<-summary[,-1]
#summary$Tencent<-as.numeric(as.character(summary$Tencent))
#summary$EA<-as.numeric(as.character(summary$EA))
for(i in 2:ncol(summary)-1){
summary[,i]<-as.numeric(as.character(summary[,i]))
}
colnames(summary)[1]<-"ActivisionBlizzard"
colnames(summary)[2]<-"Apple"
colnames(summary)[3]<-"BandaiNamco"
colnames(summary)[4]<-"DeNA"
colnames(summary)[9]<-"Microsoft"
colnames(summary)[14]<-"Squenix"
colnames(summary)[15]<-"TakeTwo"
print(summary)## ActivisionBlizzard Apple BandaiNamco DeNA Disney EA Facebook Konami
## 2 1455 1254 484 236 230 1308 181 350
## 3 1570 1349 575 233 218 1271 197 224
## 4 1172 366 464 312 179 1368 186 473
## 5 1075 350 329 476 196 955 192 264
## 6 841 390 398 506 191 711 176 345
## 7 1768 472 397 547 291 922 256 322
## 8 1324 522 440 450 194 1209 213 341
## 9 1050 565 321 429 183 949 214 191
## 10 691 603 354 383 396 695 218 235
## 11 1518 683 379 347 403 808 241 257
## 12 1111 769 357 287 268 1123 237 273
## 13 970 859 336 238 266 1214 234 171
## 14 753 897 288 225 362 990 246 202
## 15 1575 947 323 237 384 1126 257 195
## 16 1278 993 461 248 235 1185 226 241
## 17 1044 1096 426 235 208 1203 215 198
## 18 990 1150 334 229 347 815 202 232
## 19 1353 1193 446 227 390 1070 204 346
## Microsoft NetEase Nexon Nintendo Sega Squenix TakeTwo Tencent Ubisoft
## 2 1595 933 512 393 249 435 378 2646 696
## 3 1427 969 339 325 202 334 312 2652 155
## 4 809 289 351 498 201 210 148 851 212
## 5 890 316 264 491 136 131 226 890 173
## 6 973 325 281 613 180 225 273 927 196
## 7 1885 317 357 1172 224 312 416 958 1059
## 8 1583 326 423 385 195 302 300 1225 241
## 9 1039 337 349 439 151 110 143 1253 105
## 10 989 344 380 555 172 182 148 1390 300
## 11 1265 427 328 1013 218 247 1864 1398 717
## 12 1228 347 397 361 186 318 195 1673 235
## 13 970 377 308 333 166 196 125 1785 437
## 14 1017 401 381 430 199 170 126 1824 151
## 15 1428 462 359 969 272 265 531 1928 983
## 16 1416 501 432 400 219 303 300 2053 185
## 17 1463 621 355 377 169 252 275 2000 105
## 18 1517 820 414 458 167 248 347 2210 121
## 19 2013 849 381 706 231 415 414 2463 612
## newDate
## 2 2016-02-01
## 3 2016-05-01
## 4 2012-02-01
## 5 2012-05-01
## 6 2012-08-01
## 7 2012-11-01
## 8 2013-02-01
## 9 2013-05-01
## 10 2013-08-01
## 11 2013-11-01
## 12 2014-02-01
## 13 2014-05-01
## 14 2014-08-01
## 15 2014-11-01
## 16 2015-02-01
## 17 2015-05-01
## 18 2015-08-01
## 19 2015-11-01str(summary)## 'data.frame': 18 obs. of 18 variables:
## $ ActivisionBlizzard: num 1455 1570 1172 1075 841 ...
## $ Apple : num 1254 1349 366 350 390 ...
## $ BandaiNamco : num 484 575 464 329 398 397 440 321 354 379 ...
## $ DeNA : num 236 233 312 476 506 547 450 429 383 347 ...
## $ Disney : num 230 218 179 196 191 291 194 183 396 403 ...
## $ EA : num 1308 1271 1368 955 711 ...
## $ Facebook : num 181 197 186 192 176 256 213 214 218 241 ...
## $ Konami : num 350 224 473 264 345 322 341 191 235 257 ...
## $ Microsoft : num 1595 1427 809 890 973 ...
## $ NetEase : num 933 969 289 316 325 317 326 337 344 427 ...
## $ Nexon : num 512 339 351 264 281 357 423 349 380 328 ...
## $ Nintendo : num 393 325 498 491 613 ...
## $ Sega : num 249 202 201 136 180 224 195 151 172 218 ...
## $ Squenix : num 435 334 210 131 225 312 302 110 182 247 ...
## $ TakeTwo : num 378 312 148 226 273 ...
## $ Tencent : num 2646 2652 851 890 927 ...
## $ Ubisoft : num 696 155 212 173 196 ...
## $ newDate : Date, format: "2016-02-01" "2016-05-01" ...colnames(summary)## [1] "ActivisionBlizzard" "Apple" "BandaiNamco"
## [4] "DeNA" "Disney" "EA"
## [7] "Facebook" "Konami" "Microsoft"
## [10] "NetEase" "Nexon" "Nintendo"
## [13] "Sega" "Squenix" "TakeTwo"
## [16] "Tencent" "Ubisoft" "newDate"Plots
#v<-colnames(summary)
#v<-v[-length(v)]
#print(v)
#colourCount = length(unique(v))
#getPalette = colorRampPalette(brewer.pal(9, "Set1"))
g1<-ggplot(data=summary,aes(newDate))+geom_point(aes(y=EA,color="EA"),size=4)+geom_point(aes(y=Tencent,color="Tencent"),size=4)+ labs(colour='Companies') + xlab("Date") + ylab("Revenues (M $US)") + theme(legend.position='top') +ylim(0,3000)
print(g1)
g11<-ggplot(data=summary,aes(newDate))+
geom_point(aes(y=EA,color="EA"),size=2)+geom_point(aes(y=Microsoft,color="Microsoft"),size=2)+
geom_point(aes(y=Tencent,color="Tencent"),size=2)+geom_point(aes(y=Konami,color="Konami"),size=2)+
geom_point(aes(y=Squenix,color="Squenix"),size=2)+geom_point(aes(y=TakeTwo,color="TakeTwo"),size=2)+
geom_point(aes(y=NetEase,color="NetEase"),size=2)+geom_point(aes(y=Nexon,color="Nexon"),size=2)+
geom_point(aes(y=DeNA,color="DeNA"),size=2)+geom_point(aes(y=Nintendo,color="Nintendo"),size=2)+
geom_point(aes(y=Disney,color="Disney"),size=2)+geom_point(aes(y=Sega,color="Sega"),size=2)+
geom_point(aes(y=Facebook,color="Facebook"),size=2)+geom_point(aes(y=Ubisoft,color="Ubisoft"),size=2)+
geom_point(aes(y=Apple,color="Apple"),size=2)+geom_point(aes(y=BandaiNamco,color="BandaiNamco"),size=2)+
geom_point(aes(y=ActivisionBlizzard,color="ActivisionBlizzard"),size=2)+
labs(colour='Companies') + xlab("Date") + ylab("Revenues (M $US)") + theme(legend.position='top') +ylim(0,3000)
print(g11)
g2<-ggplot(data=summary,aes(x=newDate,y=Tencent))+ geom_point(size=4) + labs(colour='Companies') + xlab("Date") + ylab("Revenues (M $US)") + theme(legend.position='top') +ylim(0,3000) +geom_smooth(method="lm")
g3<-ggplot(data=summary,aes(x=newDate,y=EA))+ geom_point(size=4) + labs(colour='Companies') + xlab("Date") + ylab("Revenues (M $US)") + theme(legend.position='top') +ylim(0,3000) +geom_smooth(method="lm")
grid.arrange(g2,g3,ncol=2)
Acknoledgment Data available for free use from NEWZOO 
History :
- version 1 : initial commit