Video Games Sales Analysis
Vibhu Dagar
1.Introduction
Video games are popular all over the world. They are enjoyed by all ages. Video game industry is huge and the spending on video games per year is huge too. Sales of different types of games vary widely between countries due to local preferences. According to the market research firm SuperData, as of May 2015, the global games market was worth USD 74.2 billion. By region, North America accounted for 23.6 billion dollars, Asia for 23.1 billion dollars, Europe for 22.1 billion dollars and South America for 4.5 billion dollars. There are different genres, publisher and platforms for video games. This project relates to the sales of these video games based on different regions and analyzes the sales. Also I have analyzed which genre, platform or publisher is the most popular and has maximum number of sales.
2.Overview of the project
In this the main goal was to analyze the sales of video games in different regions. The regions are North America, Europe, Japan, other countries(comined) and then the global sales(total of all the regions). The main idea was to visualize the sales for different genres, publishers and platforms. This would give the basic idea about the most popular genres, publishers and platforms amongst all. Also analyzing the effect of genres on sales in different regions.
3.Data
For this project the data was collected from Kaggle(www.kaggle.com). This data gives us the idea about the sales of video games in different regions of the world. The distribution is with respect to genres, publishers and platforms.
Name: Name of the video game
Platform: Platform on which the game was released or is playable
Year: Year in which the game was released
Genre: Genre the game belongs to
Publisher: Name of the publisher who created the game
NA_Sales: Sales in North America
EU_Sales: Sales in Europe
JP_Sales: Sales in Japan
Other_Sales: Sales in other countries
Global_Sales: Global Sales
vgsales <- read.csv("C:/Program Files/RStudio/files/vgsales.csv")
summary(vgsales)## Rank Name Platform
## Min. : 1 Need for Speed: Most Wanted: 12 DS :2163
## 1st Qu.: 4151 FIFA 14 : 9 PS2 :2161
## Median : 8300 LEGO Marvel Super Heroes : 9 PS3 :1329
## Mean : 8301 Madden NFL 07 : 9 Wii :1325
## 3rd Qu.:12450 Ratatouille : 9 X360 :1265
## Max. :16600 Angry Birds Star Wars : 8 PSP :1213
## (Other) :16542 (Other):7142
## Year Genre Publisher
## 2009 :1431 Action :3316 Electronic Arts : 1351
## 2008 :1428 Sports :2346 Activision : 975
## 2010 :1259 Misc :1739 Namco Bandai Games : 932
## 2007 :1202 Role-Playing:1488 Ubisoft : 921
## 2011 :1139 Shooter :1310 Konami Digital Entertainment: 832
## 2006 :1008 Adventure :1286 THQ : 715
## (Other):9131 (Other) :5113 (Other) :10872
## NA_Sales EU_Sales JP_Sales Other_Sales
## Min. : 0.0000 Min. : 0.0000 Min. : 0.00000 Min. : 0.00000
## 1st Qu.: 0.0000 1st Qu.: 0.0000 1st Qu.: 0.00000 1st Qu.: 0.00000
## Median : 0.0800 Median : 0.0200 Median : 0.00000 Median : 0.01000
## Mean : 0.2647 Mean : 0.1467 Mean : 0.07778 Mean : 0.04806
## 3rd Qu.: 0.2400 3rd Qu.: 0.1100 3rd Qu.: 0.04000 3rd Qu.: 0.04000
## Max. :41.4900 Max. :29.0200 Max. :10.22000 Max. :10.57000
##
## Global_Sales
## Min. : 0.0100
## 1st Qu.: 0.0600
## Median : 0.1700
## Mean : 0.5374
## 3rd Qu.: 0.4700
## Max. :82.7400
## library(psych)
describe(vgsales)## vars n mean sd median trimmed mad min
## Rank 1 16598 8300.61 4791.85 8300.50 8300.56 6152.05 1.00
## Name* 2 16598 5795.86 3324.01 5864.50 5810.22 4270.63 1.00
## Platform* 3 16598 16.71 8.29 17.00 16.67 10.38 1.00
## Year* 4 16598 27.61 6.00 28.00 28.00 5.93 1.00
## Genre* 5 16598 5.93 3.76 6.00 5.86 5.93 1.00
## Publisher* 6 16598 299.40 181.98 329.00 303.97 272.80 1.00
## NA_Sales 7 16598 0.26 0.82 0.08 0.13 0.12 0.00
## EU_Sales 8 16598 0.15 0.51 0.02 0.06 0.03 0.00
## JP_Sales 9 16598 0.08 0.31 0.00 0.02 0.00 0.00
## Other_Sales 10 16598 0.05 0.19 0.01 0.02 0.01 0.00
## Global_Sales 11 16598 0.54 1.56 0.17 0.27 0.21 0.01
## max range skew kurtosis se
## Rank 16600.00 16599.00 0.00 -1.20 37.19
## Name* 11493.00 11492.00 -0.03 -1.21 25.80
## Platform* 31.00 30.00 -0.05 -1.00 0.06
## Year* 40.00 39.00 -0.86 1.68 0.05
## Genre* 12.00 11.00 0.07 -1.43 0.03
## Publisher* 579.00 578.00 -0.15 -1.40 1.41
## NA_Sales 41.49 41.49 18.80 648.86 0.01
## EU_Sales 29.02 29.02 18.87 755.71 0.00
## JP_Sales 10.22 10.22 11.20 194.15 0.00
## Other_Sales 10.57 10.57 24.23 1024.92 0.00
## Global_Sales 82.74 82.73 17.40 603.68 0.01==> Distribution of games with respect to consoles:
plot(vgsales$Platform, main = "Platform based distribution", xlab = "Platform", ylab = "No. of games", col="cyan")
==> Distribution of games with respect to genres:
plot(vgsales$Genre, main = "Genre based distribution", xlab = "Genre", ylab = "No. of games", col="cyan")
[ Here we see that the Action genre is the most popular amongst all ]
==> Distribution of games with respect to publishers:
plot(vgsales$Publisher, main = "Publisher based distribution", xlab = "Publisher", ylab = "No. of games")
==> Hypothesis: There is no significant change in NA_Sales with respect to genre.
[ In the following data the ones having p-value<0.05 do not have a significant change but the rest change the sales significantly.]
fit <- lm( NA_Sales ~ Genre , data = vgsales)
summary(fit)##
## Call:
## lm(formula = NA_Sales ~ Genre, data = vgsales)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.505 -0.236 -0.153 -0.015 41.199
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.264726 0.014081 18.801 < 2e-16 ***
## GenreAdventure -0.182455 0.026636 -6.850 7.65e-12 ***
## GenreFighting -0.001058 0.031202 -0.034 0.9729
## GenreMisc -0.028820 0.024007 -1.200 0.2300
## GenrePlatform 0.239846 0.030664 7.822 5.53e-15 ***
## GenrePuzzle -0.052045 0.036440 -1.428 0.1532
## GenreRacing 0.023041 0.026919 0.856 0.3921
## GenreRole-Playing -0.044779 0.025300 -1.770 0.0768 .
## GenreShooter 0.180007 0.026460 6.803 1.06e-11 ***
## GenreSimulation -0.053295 0.030928 -1.723 0.0849 .
## GenreSports 0.026557 0.021875 1.214 0.2247
## GenreStrategy -0.163845 0.034113 -4.803 1.58e-06 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.8108 on 16586 degrees of freedom
## Multiple R-squared: 0.01493, Adjusted R-squared: 0.01428
## F-statistic: 22.86 on 11 and 16586 DF, p-value: < 2.2e-16==> Hypothesis: There is no significant change in EU_Sales with respect to genre.
[ In the following data the ones having p-value<0.05 do not have a significant change but the rest change the sales significantly.]
fit <- lm( EU_Sales ~ Genre , data = vgsales)
summary(fit)##
## Call:
## lm(formula = EU_Sales ~ Genre, data = vgsales)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.2391 -0.1409 -0.1064 -0.0299 28.8594
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.158323 0.008736 18.123 < 2e-16 ***
## GenreAdventure -0.108455 0.016526 -6.563 5.44e-11 ***
## GenreFighting -0.038842 0.019358 -2.006 0.044821 *
## GenreMisc -0.034125 0.014894 -2.291 0.021966 *
## GenrePlatform 0.069250 0.019025 3.640 0.000273 ***
## GenrePuzzle -0.071072 0.022608 -3.144 0.001672 **
## GenreRacing 0.032541 0.016701 1.948 0.051380 .
## GenreRole-Playing -0.031939 0.015697 -2.035 0.041893 *
## GenreShooter 0.080814 0.016416 4.923 8.62e-07 ***
## GenreSimulation -0.027551 0.019189 -1.436 0.151088
## GenreSports 0.002312 0.013572 0.170 0.864742
## GenreStrategy -0.091745 0.021164 -4.335 1.47e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.5031 on 16586 degrees of freedom
## Multiple R-squared: 0.00971, Adjusted R-squared: 0.009054
## F-statistic: 14.79 on 11 and 16586 DF, p-value: < 2.2e-16==> Hypothesis: There is no significant change in JP_Sales with respect to genre.
[ In the following data the ones having p-value<0.05 do not have a significant change but the rest change the sales significantly.]
fit <- lm( JP_Sales ~ Genre , data = vgsales)
summary(fit)##
## Call:
## lm(formula = JP_Sales ~ Genre, data = vgsales)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.2368 -0.0620 -0.0482 -0.0282 9.9832
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.048236 0.005282 9.132 < 2e-16 ***
## GenreAdventure -0.007746 0.009992 -0.775 0.438225
## GenreFighting 0.054771 0.011705 4.679 2.9e-06 ***
## GenreMisc 0.013731 0.009006 1.525 0.127352
## GenrePlatform 0.099360 0.011503 8.638 < 2e-16 ***
## GenrePuzzle 0.050235 0.013670 3.675 0.000239 ***
## GenreRacing -0.002848 0.010098 -0.282 0.777958
## GenreRole-Playing 0.188532 0.009491 19.865 < 2e-16 ***
## GenreShooter -0.019014 0.009926 -1.916 0.055427 .
## GenreSimulation 0.025236 0.011602 2.175 0.029635 *
## GenreSports 0.009467 0.008206 1.154 0.248659
## GenreStrategy 0.024393 0.012797 1.906 0.056643 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.3042 on 16586 degrees of freedom
## Multiple R-squared: 0.03352, Adjusted R-squared: 0.03288
## F-statistic: 52.29 on 11 and 16586 DF, p-value: < 2.2e-16==> Hypothesis: There is no significant change in Other_Sales with respect to genre.
[ In the following data the ones having p-value<0.05 do not have a significant change but the rest change the sales significantly.]
fit <- lm( Other_Sales ~ Genre , data = vgsales)
summary(fit)##
## Call:
## lm(formula = Other_Sales ~ Genre, data = vgsales)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.0784 -0.0475 -0.0333 -0.0067 10.5135
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.056508 0.003262 17.321 < 2e-16 ***
## GenreAdventure -0.043436 0.006172 -7.038 2.02e-12 ***
## GenreFighting -0.013253 0.007229 -1.833 0.066783 .
## GenreMisc -0.013196 0.005562 -2.372 0.017686 *
## GenrePlatform 0.001720 0.007105 0.242 0.808696
## GenrePuzzle -0.034944 0.008443 -4.139 3.51e-05 ***
## GenreRacing 0.005358 0.006237 0.859 0.390349
## GenreRole-Playing -0.016447 0.005862 -2.806 0.005025 **
## GenreShooter 0.021881 0.006131 3.569 0.000359 ***
## GenreSimulation -0.020153 0.007166 -2.812 0.004925 **
## GenreSports 0.001024 0.005068 0.202 0.839867
## GenreStrategy -0.039826 0.007904 -5.039 4.73e-07 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.1879 on 16586 degrees of freedom
## Multiple R-squared: 0.008315, Adjusted R-squared: 0.007657
## F-statistic: 12.64 on 11 and 16586 DF, p-value: < 2.2e-16==> Hypothesis: There is no significant change in Global_Sales with respect to genre.
[ In the following data the ones having p-value<0.05 do not have a significant change but the rest change the sales significantly.]
fit <- lm( Global_Sales ~ Genre , data = vgsales)
summary(fit)##
## Call:
## lm(formula = Global_Sales ~ Genre, data = vgsales)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.928 -0.458 -0.307 -0.037 82.173
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.528100 0.026851 19.668 < 2e-16 ***
## GenreAdventure -0.342221 0.050795 -6.737 1.67e-11 ***
## GenreFighting 0.001275 0.059501 0.021 0.9829
## GenreMisc -0.062338 0.045780 -1.362 0.1733
## GenrePlatform 0.410241 0.058476 7.016 2.38e-12 ***
## GenrePuzzle -0.107224 0.069491 -1.543 0.1229
## GenreRacing 0.058001 0.051334 1.130 0.2585
## GenreRole-Playing 0.095132 0.048247 1.972 0.0486 *
## GenreShooter 0.263785 0.050458 5.228 1.74e-07 ***
## GenreSimulation -0.075736 0.058980 -1.284 0.1991
## GenreSports 0.039219 0.041715 0.940 0.3471
## GenreStrategy -0.270949 0.065052 -4.165 3.13e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.546 on 16586 degrees of freedom
## Multiple R-squared: 0.01194, Adjusted R-squared: 0.01128
## F-statistic: 18.22 on 11 and 16586 DF, p-value: < 2.2e-164.Results
From the above vizualizations we can clearly say that DC and Play Station are the most popular platforms amongst all followed by xbox. Action genre is the most popular genre of all and is followed by sports and fighting respectively. We can also see that Daito is the most popular followed by TYO and Miwasa respectively.
From the above tests we can say that the genres less popular cause significant change in sales in all regions as compared to the ones that are more popular.
5.Conclusion
By the above data we can say that action games on DC or playstation for that matter are the most popular and are the ones responsible for maximum sales all over the globe. Also as these games are so abundant and popular(ranking wise), variation in the sales of one or two such games would not cause significant change in the overall sales.
6.References
Kaggle (https://www.kaggle.com/datasets)
Wikipedia (https://en.wikipedia.org/wiki/Video_game)