1 University at Buffalo, Industrial and Systems Engineering IE322 Analytics and Computing for Industrial Engineers Lab#2 Fall 2022 Data Preprocessing and Linear Regression (This is an...

-
-
-
2022-10-21
Importing Data
library(readr)
salesdata <- read_csv("salesdata.csv", col_types = cols(`Order Date` = col_date(format = "%m/%d/%Y")))
View(salesdata)
First look of the Data
skimr::skim(salesdata)
Data summary
    Name
    salesdata
    Number of rows
    670
    Number of columns
    13
    _______________________
    
    Column type frequency:
    
    character
    7
    Date
    1
    numeric
    5
    ________________________
    
    Group variables
    None
Variable type: character
    skim_variable
    n_missing
    complete_rate
    min
    max
    empty
    n_unique
    whitespace
    Order Priority
    2
    1
    3
    13
    0
    6
    0
    Ship Mode
    2
    1
    11
    14
    0
    3
    0
    Customer Name
    1
    1
    8
    18
    0
    85
    0
    Customer Segment
    0
    1
    8
    14
    0
    4
    0
    Product Category
    0
    1
    9
    15
    0
    3
    0
    company
    0
    1
    1
    1
    0
    1
    0
    Store
    0
    1
    1
    1
    0
    4
    0
Variable type: Date
    skim_variable
    n_missing
    complete_rate
    min
    max
    median
    n_unique
    Order Date
    0
    1
    2009-01-05
    2012-12-21
    2010-12-30
    395
Variable type: numeric
    skim_variable
    n_missing
    complete_rate
    mean
    sd
    p0
    p25
    p50
    p75
    p100
    hist
    Order Quantity
    0
    1
    26.08
    17.21
    -10.00
    12.00
    25.00
    39.00
    101.00
    ▅▇▆▁▁
    Sales
    1
    1
    1915.47
    3568.45
    4.99
    166.76
    497.14
    1773.86
    26133.39
    ▇▁▁▁▁
    Profit
    2
    1
    193.54
    1205.74
    -11053.60
    -82.33
    3.89
    161.82
    10951.31
    ▁▁▇▁▁
    Unit Price
    2
    1
    95.42
    242.49
    1.26
    7.25
    22.78
    99.99
    3499.99
    ▇▁▁▁▁
    Shipping Cost
    0
    1
    13.76
    18.04
    0.49
    3.14
    6.50
    15.10
    147.12
    ▇▁▁▁▁
There are missing values in sales, profit, Order priority, ship mode, customer name and unit price.
## removing all missing values records
library(imputeTS)
## Registered S3 method overwritten by 'quantmod':
## method from
## as.zoo.data.frame zoo
S1<-na.omit(salesdata)
#replacing categorical variable missing values with mode
#mode function
library(dplyr)
##
## 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
calc_mode <- function(x) {
u <- unique(x)
tab <- tabulate(match(x, u))
u[tab == max(tab)]
}
library(tidyverse)
## ── Attaching packages
## ───────────────────────────────────────
## tidyverse 1.3.2 ──
## ✔ ggplot2 3.3.6 ✔ purrr 0.3.4
## ✔ tibble 3.1.8 ✔ stringr 1.4.1
## ✔ tidyr 1.2.1 ✔ forcats 0.5.2
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()
S2<-salesdata %>%
mutate(`Order Priority` = if_else(is.na(`Order Priority`),
calc_mode(`Order Priority`),
`Order Priority`))%>%
mutate(`Ship Mode` = if_else(is.na(`Ship Mode`),
calc_mode(`Ship Mode`),
`Ship Mode`))%>%
mutate(`Customer Name` = if_else(is.na(`Customer Name`),
calc_mode(`Customer Name`),
`Customer Name`))
#skimr::skim(S2)
S2<-na_mean(S2)
#skimr::skim(S2)
#Correletion between Sales and profit in S1
S1_c<-S1[,c("Sales","Profit")]
cor(S1_c)
## Sales Profit
## Sales 1.000000 0.480134
## Profit 0.480134 1.000000
#Correletion between Sales and profit in S2
S2_c<-S2[,c("Sales","Profit")]
cor(S2_c)
## Sales Profit
## Sales 1.0000000 0.4790883
## Profit 0.4790883 1.0000000
#Removing outliers in shipping cost
IQR<-mean(salesdata$`Shipping Cost`)+ 1.5*(15.1-3.14)
boxplot(salesdata$`Shipping Cost`, plot=FALSE)$out
## [1] 35.00 68.02 35.00 69.00 49.00 74.35 49.00 39.61 57.38 64.73
## [11] 35.00 35.00 58.95 53.48 62.74 43.75 53.03 50.00 60.49 36.61
## [21] 43.32 54.11 58.92 69.64 34.20 50.00 44.55 60.00 33.60 64.20
## [31] 66.27 99.00 36.09 41.91 35.02 35.00 87.01 48.80 66.27 69.30
## [41] 69.64 35.00 35.00 52.20 54.12 39.61 70.20 51.94 70.20 49.00
## [51] 49.00 60.00 56.14 35.00 69.00 60.00 36.61 64.73 66.67 64.73
## [61] 81.98 56.14 35.00 55.24 66.27 54.92 56.20 64.73 45.00 36.09
## [71] 70.20 84.84 49.00 69.00 49.00 60.00 45.51 55.24 147.12 49.00
## [81] 64.20 51.94
S3<-S2%>%
filter(`Shipping Cost`boxplot(S3$`Shipping Cost`)
hist(salesdata$`Shipping Cost`, prob = TRUE)

lines(density(salesdata$`Shipping Cost`), col = "green")
ggplot(salesdata, aes(x=Sales,y=Profit,color=`Ship Mode`)) +
...