Walmart Sales Analysis (2010-2012)

Introduction

Dataset Info

This is the historical data that covers sales from 2010-02-05 to 2012-11-01, in the file Walmart_Store_sales. Within this file you will find the following fields:

Store - the store number
Date - the week of sales
Weekly_Sales - sales for the given store
Holiday_Flag - whether the week is a special holiday week 1 – Holiday week 0 – Non-holiday week
Temperature - Temperature on the day of sale
Fuel_Price - Cost of fuel in the region
CPI – Prevailing consumer price index
Unemployment - Prevailing unemployment rate
Holiday Events Super Bowl: 12-Feb-10, 11-Feb-11, 10-Feb-12, 8-Feb-13 Labour Day: 10-Sep-10, 9-Sep-11, 7-Sep-12, 6-Sep-13 Thanksgiving: 26-Nov-10, 25-Nov-11, 23-Nov-12, 29-Nov-13 Christmas: 31-Dec-10, 30-Dec-11, 28-Dec-12, 27-Dec-13

Source

This dataset was taking from Kaggle. You can explore it HERE.

Objectives

Data exploration
Manipulation data (only if is necessary).
Exploration Data Analysis (EDA).
Conclusions.

Data exploration

Libraries

#Importing libraries

libraries <- c("ggplot2", "dplyr", "lubridate", 
               "readr", "plotly", "corrplot", 
               "forcats", "scatterplot3d", "tidyverse")
sapply(libraries, require, character.only = TRUE)

## Loading required package: ggplot2

## Warning: package 'ggplot2' was built under R version 4.1.3

## Loading required package: dplyr

## Warning: package 'dplyr' was built under R version 4.1.3

## 
## Attaching package: 'dplyr'

## The following objects are masked from 'package:stats':
## 
##     filter, lag

## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union

## Loading required package: lubridate

## Warning: package 'lubridate' was built under R version 4.1.3

## 
## Attaching package: 'lubridate'

## The following objects are masked from 'package:base':
## 
##     date, intersect, setdiff, union

## Loading required package: readr

## Warning: package 'readr' was built under R version 4.1.3

## Loading required package: plotly

## Warning: package 'plotly' was built under R version 4.1.3

## 
## Attaching package: 'plotly'

## The following object is masked from 'package:ggplot2':
## 
##     last_plot

## The following object is masked from 'package:stats':
## 
##     filter

## The following object is masked from 'package:graphics':
## 
##     layout

## Loading required package: corrplot

## Warning: package 'corrplot' was built under R version 4.1.3

## corrplot 0.92 loaded

## Loading required package: forcats

## Warning: package 'forcats' was built under R version 4.1.3

## Loading required package: scatterplot3d

## Warning: package 'scatterplot3d' was built under R version 4.1.3

## Loading required package: tidyverse

## Warning: package 'tidyverse' was built under R version 4.1.3

## Error: package or namespace load failed for 'tidyverse' in loadNamespace(j <- i[[1L]], c(lib.loc, .libPaths()), versionCheck = vI[[j]]):
##  namespace 'purrr' 0.3.4 is already loaded, but >= 1.0.1 is required

##       ggplot2         dplyr     lubridate         readr        plotly 
##          TRUE          TRUE          TRUE          TRUE          TRUE 
##      corrplot       forcats scatterplot3d     tidyverse 
##          TRUE          TRUE          TRUE         FALSE

setwd('E:/Waltmart_dataset')

Data loading and visualization

## Data upload 

walmart_data <- read.csv('Walmart.csv')
head(walmart_data)

##   Store       Date Weekly_Sales Holiday_Flag Temperature Fuel_Price      CPI
## 1     1 05-02-2010      1643691            0       42.31      2.572 211.0964
## 2     1 12-02-2010      1641957            1       38.51      2.548 211.2422
## 3     1 19-02-2010      1611968            0       39.93      2.514 211.2891
## 4     1 26-02-2010      1409728            0       46.63      2.561 211.3196
## 5     1 05-03-2010      1554807            0       46.50      2.625 211.3501
## 6     1 12-03-2010      1439542            0       57.79      2.667 211.3806
##   Unemployment
## 1        8.106
## 2        8.106
## 3        8.106
## 4        8.106
## 5        8.106
## 6        8.106

Database structure

num_row <- nrow(walmart_data)
num_col <- ncol(walmart_data)
sprintf("The dataset has %s samples and %s features", num_row, num_col)

## [1] "The dataset has 6435 samples and 8 features"

Count null values

## Count null values 

apply(X = is.na(walmart_data), MARGIN = 2, FUN = sum)

##        Store         Date Weekly_Sales Holiday_Flag  Temperature   Fuel_Price 
##            0            0            0            0            0            0 
##          CPI Unemployment 
##            0            0

Inference: There aren’t null values.

Count unique values

## Count unique values

sort(sapply(walmart_data, function(x) length(unique(x))))

## Holiday_Flag        Store         Date Unemployment   Fuel_Price          CPI 
##            2           45          143          349          892         2145 
##  Temperature Weekly_Sales 
##         3528         6435

Data general info

## Data info 

str(walmart_data)

## 'data.frame':    6435 obs. of  8 variables:
##  $ Store       : int  1 1 1 1 1 1 1 1 1 1 ...
##  $ Date        : chr  "05-02-2010" "12-02-2010" "19-02-2010" "26-02-2010" ...
##  $ Weekly_Sales: num  1643691 1641957 1611968 1409728 1554807 ...
##  $ Holiday_Flag: int  0 1 0 0 0 0 0 0 0 0 ...
##  $ Temperature : num  42.3 38.5 39.9 46.6 46.5 ...
##  $ Fuel_Price  : num  2.57 2.55 2.51 2.56 2.62 ...
##  $ CPI         : num  211 211 211 211 211 ...
##  $ Unemployment: num  8.11 8.11 8.11 8.11 8.11 ...

Inference: Date data type is incorrect.

Statistical summary

## Statistical summary 

summary(walmart_data)

##      Store        Date            Weekly_Sales      Holiday_Flag    
##  Min.   : 1   Length:6435        Min.   : 209986   Min.   :0.00000  
##  1st Qu.:12   Class :character   1st Qu.: 553350   1st Qu.:0.00000  
##  Median :23   Mode  :character   Median : 960746   Median :0.00000  
##  Mean   :23                      Mean   :1046965   Mean   :0.06993  
##  3rd Qu.:34                      3rd Qu.:1420159   3rd Qu.:0.00000  
##  Max.   :45                      Max.   :3818686   Max.   :1.00000  
##   Temperature       Fuel_Price         CPI         Unemployment   
##  Min.   : -2.06   Min.   :2.472   Min.   :126.1   Min.   : 3.879  
##  1st Qu.: 47.46   1st Qu.:2.933   1st Qu.:131.7   1st Qu.: 6.891  
##  Median : 62.67   Median :3.445   Median :182.6   Median : 7.874  
##  Mean   : 60.66   Mean   :3.359   Mean   :171.6   Mean   : 7.999  
##  3rd Qu.: 74.94   3rd Qu.:3.735   3rd Qu.:212.7   3rd Qu.: 8.622  
##  Max.   :100.14   Max.   :4.468   Max.   :227.2   Max.   :14.313

Manipulating data

## Date data type

class(walmart_data$Date)

## [1] "character"

Inference: It’s necesary change to type data.

## Change data type

walmart_data<-walmart_data %>%
  mutate(Date=dmy(Date)) 

## Verification of change 

sprintf("The data type of the Date variable is: %s", class(walmart_data$Date) )

## [1] "The data type of the Date variable is: Date"

## Classifying fuel prices

walmart_data <-
  walmart_data %>% mutate(Sts_Fuel_Price = ifelse(Fuel_Price < mean(walmart_data$Fuel_Price), "Low", "High"))

## Classifying temperature

walmart_data <-
  walmart_data %>% mutate(Sts_Temperature = case_when( 
    Temperature <= quantile(walmart_data$Temperature, c(0.25)) ~ "Cool",
    Temperature > quantile(walmart_data$Temperature, c(0.50)) & 
     Temperature <= quantile(walmart_data$Temperature, c(0.75)) ~ "Cold", 
    T ~ "Hot"))

## Classifying CPI

walmart_data <- 
  walmart_data %>% mutate(Sts_CPI = ifelse(
    CPI > mean(walmart_data$CPI), "High", "Low"
  ))

## Filtering day, month and year from Date

Weekday <- day(walmart_data$Date)
Month <- month(walmart_data$Date)
Year <- year(walmart_data$Date)

## Add Day column 

walmart_data <- walmart_data %>%
  cbind(Weekday)

## Add Month column

walmart_data <- walmart_data %>%
  cbind(Month)

## Add Year column 
walmart_data <- walmart_data %>%
  cbind(Year)

Exploration Data Analysis (EDA)

Response variable: Weekly_Sales.

# Target variable distribution

ggplot(walmart_data, aes(x = Weekly_Sales)) +
  
geom_histogram(aes(y=..density..),
               color = 'black', 
               fill = "#24F1B0", 
               bins = 30) +
  
  geom_density(alpha = 0.2, fill="white") +
  
ggtitle('Target Variable Distribution') +
labs(y = 'Frequency')

## Warning: The dot-dot notation (`..density..`) was deprecated in ggplot2 3.4.0.
## i Please use `after_stat(density)` instead.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.

Inference: The Target Variable seems to be be normally distributed, averaging around 20 units.

## Select numeric features

walmart_numeric_features <- walmart_data %>% 
  select(Store, Weekly_Sales, Holiday_Flag, Temperature,
         Fuel_Price, CPI, Unemployment, Weekday, Month, Year)

## Correlation between numeric variables 

round(abs(cor(walmart_numeric_features)), 3)

##              Store Weekly_Sales Holiday_Flag Temperature Fuel_Price   CPI
## Store        1.000        0.335        0.000       0.023      0.060 0.209
## Weekly_Sales 0.335        1.000        0.037       0.064      0.009 0.073
## Holiday_Flag 0.000        0.037        1.000       0.155      0.078 0.002
## Temperature  0.023        0.064        0.155       1.000      0.145 0.177
## Fuel_Price   0.060        0.009        0.078       0.145      1.000 0.171
## CPI          0.209        0.073        0.002       0.177      0.171 1.000
## Unemployment 0.224        0.106        0.011       0.101      0.035 0.302
## Weekday      0.000        0.017        0.045       0.027      0.028 0.003
## Month        0.000        0.076        0.123       0.236      0.042 0.005
## Year         0.000        0.018        0.057       0.064      0.779 0.075
##              Unemployment Weekday Month  Year
## Store               0.224   0.000 0.000 0.000
## Weekly_Sales        0.106   0.017 0.076 0.018
## Holiday_Flag        0.011   0.045 0.123 0.057
## Temperature         0.101   0.027 0.236 0.064
## Fuel_Price          0.035   0.028 0.042 0.779
## CPI                 0.302   0.003 0.005 0.075
## Unemployment        1.000   0.004 0.013 0.242
## Weekday             0.004   1.000 0.015 0.006
## Month               0.013   0.015 1.000 0.194
## Year                0.242   0.006 0.194 1.000

## Heatmap 

#heatmap(abs(cor(walmart_numeric_features)), Rowv = NA, Colv = NA, revC = T)

## Corrplot

corrplot(cor(walmart_numeric_features), method='shade', shade.col= NA,
         tl.col = "black", tl.srt = 50, addCoef.col = 'black')

Inference: The correlation between the dependent variable ( Weekly_Sales) and the dependent variables is relatively low.

# Weekly sales for year

walmart_data%>%
  group_by(Date)%>%
  
  summarise(Weekly_Sales = mean(Weekly_Sales, na.rm=T)) %>%
  
  ggplot(aes(Date, Weekly_Sales)) +
  
  geom_point(aes(color=Weekly_Sales>1200000),
             show.legend = F) +
  geom_line(color='grey') +
  
  labs(title = 'Weekly Sales for year',
       y='Weekly sales',
       x='Date') +

    theme_bw()

Inference: some dates have extreme high weekly sales.

# Holiday and Non-Holiday sales

holidays <-function(x){
  cut(x,
      breaks = ymd(20100101, 20101126, 20101127, 20101224,20101225,
                   20111125, 20111126, 20111223, 20111224,20121123,
                   20121124, 20121221, 20121222, 20130101),
      
      labels = c('Normal Day','Thanksgiving', 
                 'Normal Day', 'A week before Christmas',
                 'Normal Day', 'Thanksgiving', 
                 'Normal Day', 'A week before Christmas',
                 'Normal Day', 'Thanksgiving', 
                 'Normal Day', 'A week before Christmas',
                 'Normal Day'))
}

walmart_data <- walmart_data %>%
  mutate(festival = holidays(Date))

walmart_data %>%
  group_by(Date)%>%
  
  mutate(Weekly_Sales = mean(Weekly_Sales, na.rm=T)) %>%
  
  ggplot(aes(Date, Weekly_Sales)) +
  
  geom_point(aes(color = festival, 
                 shape = festival),
             size = 2)+
  geom_line(group=1, color='grey')

Inference : festival will lead to high weekly sales, especially a week before the Chritmas and the Thanksgiving.

## Annual Sales 
Annual_Sales <- 
  aggregate(Weekly_Sales ~ Year, walmart_data, sum)

Annual_Sales <- Annual_Sales[order(-Annual_Sales$Weekly_Sales), ]

print(Annual_Sales)

##   Year Weekly_Sales
## 2 2011   2448200007
## 1 2010   2288886120
## 3 2012   2000132859

## Annual sales pie plot

sales <- Annual_Sales$Weekly_Sales
labels <- c('2010', '2011', '2012' )
porcent <- round(sales/sum(sales)*100)
labels <- paste(labels, porcent, "%", sep = " ")

pie(sales,
    labels = labels,
    col = rainbow(length(labels)),
    main = "Annual Sales")

legend("topright", c("2010", "2011", "2012"),
       cex = 0.8,
       fill = rainbow(length(labels)))

# Scatterplot: Year-weekly sales

walmart_data %>%
  
  ggplot(aes(x = Year,
             y = Weekly_Sales)) +
  geom_point(alpha = 0.1, 
             colour = 'blue') +
  
  labs(title = 
         'Year / Weekly_Sales Scatterplot',
       x='Year',
       y='Weekly_Sales')

Inferece:

The years 2010 and 2011 reported higher earnings.
The years 2010 and 2011 present atypical data.

Annual Stores Sales

## Top 10 stores

Store_Sales <- 
  aggregate(Weekly_Sales ~ Store, walmart_data, sum)

Top_10_Stores <- head(Store_Sales[order(-Store_Sales$Weekly_Sales), ], 10)

print(Top_10_Stores)

##    Store Weekly_Sales
## 20    20    301397792
## 4      4    299543953
## 14    14    288999911
## 13    13    286517704
## 2      2    275382441
## 10    10    271617714
## 27    27    253855917
## 6      6    223756131
## 1      1    222402809
## 39    39    207445542

## Barplot (Relative Frequency)

barplot(prop.table(Top_10_Stores$Weekly_Sales) * 100, 
        names.arg = c(Top_10_Stores$Store),
        col = 'blue', 
        main = "Top 10 best-selling stores ", 
        xlab = "Stores",
        ylab = "Weekly Sales")

Inference: The 20th store reported more annual sales.

Holiday Flag

# Holiday flags classes

walmart_data %>% 
  count(Holiday_Flag) %>%
  plot_ly( x = ~Holiday_Flag, 
           y= ~n, 
           type = 'bar',
           text = ~n,
           marker = list(color = 'rgb(158,202,225)',
                         line = list(color = 'rgb(8,48,107)',
                                     width = 0.5)))%>%
  layout(title = "Holiday Flag Classes",
         xaxis = list(title = "Holiday_Flag"),
         yaxis = list(title = "Frequency"))

# Boxplot holiday flags-Weekly sales

walmart_data %>% 
  ggplot(aes(as_factor(Holiday_Flag), Weekly_Sales)) +
  
  geom_boxplot() + 
  labs(title = "Boxplot Holiday Flag",
       x = "Holiday Flag", 
       y = "Weekly Sales") + 
 
   theme_bw()

## Scatterplot Weekly Sales / Holiday Flag

walmart_data %>% 
  ggplot(aes(x = Holiday_Flag, 
             y = Weekly_Sales)) +
  geom_point(alpha = 0.1, 
             colour = 'blue') +
  
  labs(title = 
         'Weekly sales / Holiday Flag Scatterplot',
       y='Weekly sales',
       x='Holiday Flag')

# Holiday flags sales

Sales_Holidays <-
  aggregate(Weekly_Sales ~ Holiday_Flag, walmart_data, sum)

Sales_Holidays <-
  Sales_Holidays[order(-Sales_Holidays$Holiday_Flag),]

print(Sales_Holidays)

##   Holiday_Flag Weekly_Sales
## 2            1    505299552
## 1            0   6231919436

Inference:

There are more sales in Non-Holiday week.
However, there is more revenues in Holiday week.
There are extreme value in Non-holiday week.

Fuel Price

# Weekly Sales / Fuel_Price

Sales_vs_Fuel_Price <- 
  aggregate(Weekly_Sales ~ Sts_Fuel_Price, walmart_data, sum)

print(Sales_vs_Fuel_Price)

##   Sts_Fuel_Price Weekly_Sales
## 1           High   3628932238
## 2            Low   3108286749

## Fuel_Price distribution

hist_fp <- ggplot(walmart_data, aes(x=Fuel_Price)) +
  
  geom_histogram(aes(y=..density..),
                 position = "identity",
                 binwidth = 0.10,
                 colour = 'black', 
                 fill = "lightblue") +
  
  geom_density(alpha = 0.2, fill="white") 

box_fp <- ggplot(walmart_data, aes(x="", y=Fuel_Price)) + 
  
  geom_boxplot(fill='lightblue', color ='black') + 
  
  coord_flip() + 
  theme_bw() + 
  xlab("") + 
  theme(axis.text.y=element_blank(),
        axis.ticks.y=element_blank())

subplot(hist_fp, box_fp, nrows = 2) %>%
  layout( title = 'Fuel Price distribution',
         lot_bgcolor='#e5ecf6', 
         xaxis = list( 
           zerolinecolor = '#ffff', 
           zerolinewidth = 2, 
           gridcolor = 'ffff'), 
         yaxis = list( 
           zerolinecolor = '#ffff', 
           zerolinewidth = 2, 
           gridcolor = 'ffff'))

## Warning: 'layout' objects don't have these attributes: 'lot_bgcolor'
## Valid attributes include:
## '_deprecated', 'activeshape', 'annotations', 'autosize', 'autotypenumbers', 'calendar', 'clickmode', 'coloraxis', 'colorscale', 'colorway', 'computed', 'datarevision', 'dragmode', 'editrevision', 'editType', 'font', 'geo', 'grid', 'height', 'hidesources', 'hoverdistance', 'hoverlabel', 'hovermode', 'images', 'legend', 'mapbox', 'margin', 'meta', 'metasrc', 'modebar', 'newshape', 'paper_bgcolor', 'plot_bgcolor', 'polar', 'scene', 'selectdirection', 'selectionrevision', 'separators', 'shapes', 'showlegend', 'sliders', 'smith', 'spikedistance', 'template', 'ternary', 'title', 'transition', 'uirevision', 'uniformtext', 'updatemenus', 'width', 'xaxis', 'yaxis', 'barmode', 'bargap', 'mapType'

# Scatterplot: weekly sales / fuel price 

walmart_data %>% 
  ggplot(aes(x = Fuel_Price, 
             y = Weekly_Sales)) +
  geom_point(alpha = 0.1, 
             colour = 'blue') +
  
  labs(title = 
         'Weekly sales / Fuel Price Scatterplot',
       y='Weekly sales',
       x='Fuel Price')

Inference: There are more people to shop in specific fuel price.

CPI

# Total Sales and CPI

Sales_vs_CPI <- 
  aggregate(Weekly_Sales ~ Sts_CPI, walmart_data, sum)

print(Sales_vs_CPI)

##   Sts_CPI Weekly_Sales
## 1    High   3330247700
## 2     Low   3406971287

## Histogram: CPI

hist_cpi <- ggplot(walmart_data, aes(x=CPI)) +
  
  geom_histogram(aes(y=..density..),
                 position = "identity",
                 binwidth = 2,
                 colour = 'blue', 
                 fill = "lightblue") +
  
  geom_density(alpha = 0.2, fill="white") 

box_cpi <- ggplot(walmart_data, aes(x="", y=CPI)) + 
  
  geom_boxplot(fill='lightblue', color ='black') + 
  
  coord_flip() + 
  theme_bw() + 
  xlab("") + 
  theme(axis.text.y=element_blank(),
        axis.ticks.y=element_blank())

subplot(hist_cpi, box_cpi, nrows = 2) %>%
  layout(title = 'CPI Distribution', 
         lot_bgcolor='#e5ecf6', 
         xaxis = list( 
           zerolinecolor = '#ffff', 
           zerolinewidth = 2, 
           gridcolor = 'ffff'), 
         yaxis = list( 
           zerolinecolor = '#ffff', 
           zerolinewidth = 2, 
           gridcolor = 'ffff'))

## Warning: 'layout' objects don't have these attributes: 'lot_bgcolor'
## Valid attributes include:
## '_deprecated', 'activeshape', 'annotations', 'autosize', 'autotypenumbers', 'calendar', 'clickmode', 'coloraxis', 'colorscale', 'colorway', 'computed', 'datarevision', 'dragmode', 'editrevision', 'editType', 'font', 'geo', 'grid', 'height', 'hidesources', 'hoverdistance', 'hoverlabel', 'hovermode', 'images', 'legend', 'mapbox', 'margin', 'meta', 'metasrc', 'modebar', 'newshape', 'paper_bgcolor', 'plot_bgcolor', 'polar', 'scene', 'selectdirection', 'selectionrevision', 'separators', 'shapes', 'showlegend', 'sliders', 'smith', 'spikedistance', 'template', 'ternary', 'title', 'transition', 'uirevision', 'uniformtext', 'updatemenus', 'width', 'xaxis', 'yaxis', 'barmode', 'bargap', 'mapType'

# Scatterplot: weekly sales / CPI

walmart_data %>% 
  ggplot(aes(x = CPI,
             y = Weekly_Sales)) +
  geom_point(alpha = 0.1, 
             colour = 'blue') +
  
  labs(title = 
         'Weekly sales / CPI Scatterplot',
       y='Weekly sales',
       x='CPI')

Inference: Seems that different range of CPI have same sales distributions. Inference: Walmart reported higher profits when fuel prices and CPI were high and when the temperature was cold.

Temperature

## Weekly_Sales / Temperature

Sales_vs_Temperature <- 
  aggregate(Weekly_Sales ~ Sts_Temperature, walmart_data, sum)

print(Sales_vs_Temperature)

##   Sts_Temperature Weekly_Sales
## 1            Cold   1696418667
## 2            Cool   1758497608
## 3             Hot   3282302713

## Temperature distribution

#library(htmltools)
hist_temp <- ggplot(walmart_data, 
                    aes(x=Temperature)) +
  
  geom_histogram(aes(y=..density..),
                 position = "identity",
                 bins = 30,
                 colour = 'blue', 
                 fill = "lightblue") +
  
  geom_density(alpha = 0.2, fill="white") 

box_temp <- ggplot(walmart_data, aes(x="", y=Temperature)) + 
  
  geom_boxplot(fill='lightblue', color ='black') + 
  
  coord_flip() + 
  theme_bw() + 
  xlab("") + 
  theme(axis.text.y=element_blank(),
        axis.ticks.y=element_blank())

subplot(hist_temp, box_temp, nrows = 2) %>%
  plotly::layout(title = 'Temperature Distribution',
         lot_bgcolor='#e5ecf6', 
         xaxis = list( 
           zerolinecolor = '#ffff', 
           zerolinewidth = 2, 
           gridcolor = 'ffff'), 
         yaxis = list( 
           zerolinecolor = '#ffff', 
           zerolinewidth = 2, 
           gridcolor = 'ffff'))

## Warning: 'layout' objects don't have these attributes: 'lot_bgcolor'
## Valid attributes include:
## '_deprecated', 'activeshape', 'annotations', 'autosize', 'autotypenumbers', 'calendar', 'clickmode', 'coloraxis', 'colorscale', 'colorway', 'computed', 'datarevision', 'dragmode', 'editrevision', 'editType', 'font', 'geo', 'grid', 'height', 'hidesources', 'hoverdistance', 'hoverlabel', 'hovermode', 'images', 'legend', 'mapbox', 'margin', 'meta', 'metasrc', 'modebar', 'newshape', 'paper_bgcolor', 'plot_bgcolor', 'polar', 'scene', 'selectdirection', 'selectionrevision', 'separators', 'shapes', 'showlegend', 'sliders', 'smith', 'spikedistance', 'template', 'ternary', 'title', 'transition', 'uirevision', 'uniformtext', 'updatemenus', 'width', 'xaxis', 'yaxis', 'barmode', 'bargap', 'mapType'

## Distribution Weekly_sale/Temperature

walmart_data %>% 
  ggplot(aes(x = Temperature, 
             y = Weekly_Sales)) +
  geom_point(alpha = 0.1, 
             colour = 'blue') +
  
  labs(title = 
         'Weekly sales / Temperature Correlation',
       y='Weekly sales',
       x='Temperature')

Inference: Temperature: Some people go shopping when the temperature is between 40 and 75 degrees Farenheint.

Unemployment

## Unemployment distribution 

hist_u <- ggplot(walmart_data, aes(x=Unemployment)) +
  
  geom_histogram(aes(y=..density..),
                 position = "identity",
                 binwidth = 0.10,
                 colour = 'blue', 
                 fill = "lightblue") +
  
  geom_density(alpha = 0.2, fill="white") 

box_u <- ggplot(walmart_data, aes(x="", y=Unemployment)) + 
  
  geom_boxplot(fill='lightblue', color ='black') + 
  
  coord_flip() + 
  theme_bw() + 
  xlab("") + 
  theme(axis.text.y=element_blank(),
        axis.ticks.y=element_blank())

subplot(hist_u, box_u, nrows = 2) %>%
  layout(title = 'Unemployment distribution ',
         lot_bgcolor='#e5ecf6', 
         xaxis = list( 
           zerolinecolor = '#ffff', 
           zerolinewidth = 2, 
           gridcolor = 'ffff'), 
         yaxis = list( 
           zerolinecolor = '#ffff', 
           zerolinewidth = 2, 
           gridcolor = 'ffff'))

## Warning: 'layout' objects don't have these attributes: 'lot_bgcolor'
## Valid attributes include:
## '_deprecated', 'activeshape', 'annotations', 'autosize', 'autotypenumbers', 'calendar', 'clickmode', 'coloraxis', 'colorscale', 'colorway', 'computed', 'datarevision', 'dragmode', 'editrevision', 'editType', 'font', 'geo', 'grid', 'height', 'hidesources', 'hoverdistance', 'hoverlabel', 'hovermode', 'images', 'legend', 'mapbox', 'margin', 'meta', 'metasrc', 'modebar', 'newshape', 'paper_bgcolor', 'plot_bgcolor', 'polar', 'scene', 'selectdirection', 'selectionrevision', 'separators', 'shapes', 'showlegend', 'sliders', 'smith', 'spikedistance', 'template', 'ternary', 'title', 'transition', 'uirevision', 'uniformtext', 'updatemenus', 'width', 'xaxis', 'yaxis', 'barmode', 'bargap', 'mapType'

# Scatterplot Weekly Sales / Unemployment

walmart_data %>% 
  ggplot(aes(x = Unemployment,
             y = Weekly_Sales)) +
  geom_point(alpha = 0.1, 
             colour = 'blue') +
  
  labs(title = 
         'Weekly sales / Unemployment Scatterplot',
       y='Weekly sales',
       x='Unemployment')

Inference: Lower unemployment rate has higher weekly sales.

Sales Analysis by Year.

## Filtering sales by years

Sales_2010 <- subset(walmart_data, Year == 2010)
Sales_2011 <- subset(walmart_data, Year == 2011)
Sales_2012 <- subset(walmart_data, Year == 2012)

Sales analysis 2010

## Total sales 2010

Total_Sales_2010 <- 
  aggregate(Weekly_Sales ~ Store, Sales_2010, sum)

Sales_Top5_2010 <- 
  head(Total_Sales_2010[order(-Total_Sales_2010$Weekly_Sales), ], 5)

print(Sales_Top5_2010)

##    Store Weekly_Sales
## 14    14    105462242
## 20    20    101733081
## 4      4     95680471
## 2      2     95277864
## 13    13     95272735

## Total sales per month (2010)

Total_Sales_per_Month_2010 <- 
  aggregate(Weekly_Sales ~ Month, Sales_2010, sum)
Total_Sales_per_Month_2010 <- Total_Sales_per_Month_2010[order(-Total_Sales_per_Month_2010$Weekly_Sales),]

print(Total_Sales_per_Month_2010)

##    Month Weekly_Sales
## 11    12    288760533
## 6      7    232580126
## 3      4    231412368
## 9     10    217161824
## 10    11    202853370
## 5      6    192246172
## 1      2    190332983
## 7      8    187640111
## 4      5    186710934
## 2      3    181919803
## 8      9    177267896

## Total sales per holidays (2010)

Sales_Holidays_2010 <-
  aggregate(Weekly_Sales ~ Holiday_Flag, Sales_2010, sum)

Sales_Holidays_2010 <-
  Sales_Holidays_2010[order(-Sales_Holidays_2010$Holiday_Flag),]

print(Sales_Holidays_2010)

##   Holiday_Flag Weekly_Sales
## 2            1    200224598
## 1            0   2088661523

## Total sales and fuel prices (2010)

Sales_vs_Fuel_Price_2010 <- 
  aggregate(Weekly_Sales ~ Sts_Fuel_Price, Sales_2010, sum)

print(Sales_vs_Fuel_Price_2010)

##   Sts_Fuel_Price Weekly_Sales
## 1            Low   2288886120

## Total sales and CPI (2010)

Sales_vs_CPI_2010 <- 
  aggregate(Weekly_Sales ~ Sts_CPI, Sales_2010, sum)

Sales_vs_CPI_2010 <- 
  Sales_vs_CPI_2010[order(-Sales_vs_CPI_2010$Weekly_Sales),]

print(Sales_vs_CPI_2010)

##   Sts_CPI Weekly_Sales
## 2     Low   1164241673
## 1    High   1124644447

## Total sales and temperature (2010)

Sales_vs_Temperature_2010 <- 
  aggregate(Weekly_Sales ~ Sts_Temperature, Sales_2010, sum)

Sales_vs_Temperature_2010 <-
  Sales_vs_Temperature_2010[order(-Sales_vs_Temperature_2010$Weekly_Sales),]

print(Sales_vs_Temperature_2010)

##   Sts_Temperature Weekly_Sales
## 3             Hot   1117750478
## 2            Cool    624263738
## 1            Cold    546871904

Inference:

There are more revenues in Holiday week.
The month of December was reported with more revenues.
The 14th store reported more annual sales.
Higher revenues were recorded when fuel prices and CPI were low and the temperature was hot.

Analysis Sales 2011

## Total sales (2011)

Total_Sales_2011 <- 
  aggregate(Weekly_Sales ~ Store, Sales_2011, sum)

Sales_Top_5_2011 <- 
  head(Total_Sales_2011[order(-Total_Sales_2011$Weekly_Sales), ], 5)

print(Sales_Top_5_2011)

##    Store Weekly_Sales
## 4      4    111092293
## 20    20    109837002
## 14    14    106096271
## 13    13    104537513
## 10    10     98916895

## Total sales per month (2011)

Total_Sales_per_Month_2011 <- 
  aggregate(Weekly_Sales ~ Month, Sales_2011, sum)
Total_Sales_per_Month_2011 <- Total_Sales_per_Month_2011[order(-Total_Sales_per_Month_2011$Weekly_Sales),]

print(Total_Sales_per_Month_2011)

##    Month Weekly_Sales
## 12    12    288078102
## 7      7    229911399
## 4      4    226526511
## 9      9    220847738
## 11    11    210162355
## 6      6    189773385
## 8      8    188599332
## 2      2    186331328
## 10    10    183261283
## 5      5    181648158
## 3      3    179356448
## 1      1    163703967

## Total sales per holidays (2011)

Sales_Holidays_2011 <-
  aggregate(Weekly_Sales ~ Holiday_Flag, Sales_2011, sum)

Sales_Holidays_2011 <-
  Sales_Holidays_2011[order(-Sales_Holidays_2011$Holiday_Flag),]

print(Sales_Holidays_2011)

##   Holiday_Flag Weekly_Sales
## 2            1    206735487
## 1            0   2241464521

## Total sales and Fuel prices (2011)

Sales_vs_Fuel_Price_2011 <- 
  aggregate(Weekly_Sales ~ Sts_Fuel_Price, Sales_2011, sum)

Sales_vs_Fuel_Price_2011 <- 
  Sales_vs_Fuel_Price_2011[order(-Sales_vs_Fuel_Price_2011$Weekly_Sales), ]

print(Sales_vs_Fuel_Price_2011)

##   Sts_Fuel_Price Weekly_Sales
## 1           High   1847344472
## 2            Low    600855535

## Total sales and CPI (2011)

Sales_vs_CPI_2011 <- 
  aggregate(Weekly_Sales ~ Sts_CPI, Sales_2011, sum)

Sales_vs_CPI_2011 <-
  Sales_vs_CPI_2011[order(-Sales_vs_CPI_2011$Weekly_Sales),]

print(Sales_vs_CPI_2011)

##   Sts_CPI Weekly_Sales
## 2     Low   1234010115
## 1    High   1214189892

## Total sales and temperature (2011)

Sales_vs_Temperature_2011 <- 
  aggregate(Weekly_Sales ~ Sts_Temperature, Sales_2011, sum)

Sales_vs_Temperature_2011 <-
  Sales_vs_Temperature_2011[order(-Sales_vs_Temperature_2011$Weekly_Sales),]

print(Sales_vs_Temperature_2011)

##   Sts_Temperature Weekly_Sales
## 3             Hot   1087110090
## 2            Cool    772472295
## 1            Cold    588617623

Inference:

There are more revenues in Holiday week.
The 4th store reported more annual sales.
The month of December was reported with more revenues.
Higher revenues were recorded when fuel prices was high, CPI was low and the temperature was Hot.

Analysis Sales 2012

## Total sales (2012)

Total_Sales_per_Store_2012 <- 
  aggregate(Weekly_Sales ~ Store, Sales_2012, sum)

Sales_Top5_2012 <- 
  head(Total_Sales_per_Store_2012[order(-Total_Sales_per_Store_2012$Weekly_Sales),], 5)

print(Sales_Top5_2012)

##    Store Weekly_Sales
## 4      4     92771189
## 20    20     89827709
## 13    13     86707455
## 2      2     81496695
## 10    10     78228617

## Total sales per month (2012)

Total_Sales_per_Month_2012 <- 
  aggregate(Weekly_Sales ~ Month, Sales_2012, sum)

Total_Sales_per_Month_2012 <- Total_Sales_per_Month_2012[order(-Total_Sales_per_Month_2012$Weekly_Sales),]

print(Total_Sales_per_Month_2012)

##    Month Weekly_Sales
## 6      6    240610329
## 8      8    236850766
## 3      3    231509650
## 2      2    192063580
## 4      4    188920906
## 5      5    188766479
## 7      7    187509452
## 10    10    184361680
## 9      9    180645544
## 1      1    168894472

## Total sales per holidays (2012)

Sales_Holidays_2012 <-
  aggregate(Weekly_Sales ~ Holiday_Flag, Sales_2012, sum)

Sales_Holidays_2012 <-
  Sales_Holidays_2012[order(-Sales_Holidays_2012$Holiday_Flag),]

print(Sales_Holidays_2012)

##   Holiday_Flag Weekly_Sales
## 2            1     98339467
## 1            0   1901793392

## Total sales and Fuel prices (2012)

Sales_vs_Fuel_Price_2012 <- 
  aggregate(Weekly_Sales ~ Sts_Fuel_Price, Sales_2012, sum)

Sales_vs_Fuel_Price_2010 <-  
  Sales_vs_Fuel_Price_2012[order(-Sales_vs_Fuel_Price_2012$Weekly_Sales),]

print(Sales_vs_Fuel_Price_2012)

##   Sts_Fuel_Price Weekly_Sales
## 1           High   1781587766
## 2            Low    218545093

## Total sales and CPI

Sales_vs_CPI_2012 <- 
  aggregate(Weekly_Sales ~ Sts_CPI, Sales_2012, sum)

Sales_vs_CPI_2012 <- 
  Sales_vs_CPI_2012[order(-Sales_vs_CPI_2012$Weekly_Sales),]

print(Sales_vs_CPI_2012)

##   Sts_CPI Weekly_Sales
## 2     Low   1008719499
## 1    High    991413361

## Total sales and temperature (2012)

Sales_vs_Temperature_2012 <- 
  aggregate(Weekly_Sales ~ Sts_Temperature, Sales_2012, sum)

Sales_vs_Temperature_2012 <-
  Sales_vs_Temperature_2012[order(~Sales_vs_Temperature_2012),]

## Warning in is.na(x): is.na() aplicado a un objeto que no es (lista o vector) de
## tipo 'language

print(Sales_vs_Temperature_2012)

##   Sts_Temperature Weekly_Sales
## 1            Cold    560929140
## 2            Cool    361761575

unique(Sales_2012$Month)

##  [1]  1  2  3  4  5  6  7  8  9 10

Inference:

There are more revenues in Holiday week.
The 4th store reported more annual sales.
The month of June was reported with more revenues.
Higher revenues were recorded when fuel prices was high, CPI was low and the temperature was cold.
Only the first ten months are recorded.

Conclusions

The year 2010 was the year with the highest revenues, followed by 2011.
The stores with the highest revenues are number 14, 20 and 4.
In general, holidays have the highest revenues.
There are lower revenues when the temperature is low and higher revenues when the temperature is high.
There are higher revenues when the CPI is low.
The issue of rising fuel prices does not seem to affect store revenues.

Walmart Sales Analysis (2010-2012)

Carlos Eduardo Salinas Díaz

Introduction

Dataset Info

Source

Objectives

Data exploration

Libraries

Data loading and visualization

Database structure

Count null values

Count unique values

Data general info

Statistical summary

Manipulating data

Exploration Data Analysis (EDA)

Annual Stores Sales

Holiday Flag

Fuel Price

CPI

Temperature

Unemployment

Sales Analysis by Year.

Sales analysis 2010

Analysis Sales 2011

Analysis Sales 2012

Conclusions