-
Notifications
You must be signed in to change notification settings - Fork 0
/
final v10.Rmd
144 lines (109 loc) · 4.78 KB
/
final v10.Rmd
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
---
title: "Buy Online Pick Up in Store (BOPS)"
author: "Kevin, Ram, Ryo"
date: "June 6, 2017"
output:
slidy_presentation: default
ioslides_presentation: default
---
```{r echo=FALSE, warning=FALSE, message=FALSE}
setwd("C:/Users/kevin/Dropbox/SCU/OMIS 3392/project")
# Assessing the impact of buy-online-pick up-in store strategy on online and Brick-and-mortar store sales and returns
```
```{r echo=FALSE, warning=FALSE, message=FALSE}
## Load Data Set
library(data.table)
library(readr)
library(dplyr)
library(VIF)
library(usdm)
library(AER)
library(foreign)
bops <- read_csv("C:/Users/kevin/Dropbox/SCU/OMIS 3392/project/data.txt")
bops <- as.data.table(bops)
bops2012 <- fread("C:/Users/kevin/Dropbox/SCU/OMIS 3392/project/BOPS-FY12.csv")
bops2013 <- fread("C:/Users/kevin/Dropbox/SCU/OMIS 3392/project/BOPS-FY13.csv")
# bops - original data file
# bops1 - used for returns model
# bops2 - used for sales model
# bops3 - sales model for store 2 and 6
```
```{r echo=FALSE, warning=FALSE, message=FALSE}
bops$gender[is.na(bops$gender)] <- "U"
```
```{r echo=FALSE, warning=FALSE, message=FALSE}
bops$homeowner_code[bops$homeowner_code == "R"] <- 0
bops$homeowner_code[bops$homeowner_code == "O"] <- 1
```
```{r echo=FALSE, warning=FALSE, message=FALSE}
bops$store_number <- as.factor(bops$store_number)
bops <- within(bops, store_number <- relevel(store_number, ref = 3))
```
```{r echo=FALSE, warning=FALSE, message=FALSE}
# Add column to the individual years
bops2012 <- bops2012[,-c(1,3)]
bops2012 <- cbind(bops2012, b2012 = 1)
bops2013 <- bops2013[,-c(1,3)]
bops2013 <- cbind(bops2013, b2013 = 1)
# Merge dataframes
bops <- merge(bops, bops2012, by="transaction_id", all.x=TRUE)
bops <- merge(bops, bops2013, by="transaction_id", all.x=TRUE)
# fill in NA
bops$b2012[is.na(bops$b2012)] <- 0
bops$b2013[is.na(bops$b2013)] <- 0
#merge the 2 columns
bops$bops <- bops$b2012 + bops$b2013
bops$gender <- as.factor(bops$gender)
bops$ethnic_code <- as.factor(bops$ethnic_code)
```
```{r echo=FALSE, warning=FALSE, message=FALSE}
bops2 <- bops[,-c(3,6:10,18:19, 22:23)]
```
```{r echo=FALSE, warning=FALSE, message=FALSE}
bops2$customer_id <- as.factor(bops2$customer_id)
bops2a <- bops2[which(bops2$month_index <= 24),]
bops2b <- bops2[which(bops2$month_index <= 36 & bops2$month_index >= 25),]
# determine duplicates
a = count(bops2a, customer_id)
bops2a <- merge(bops2a, a, by="customer_id", all.x=TRUE)
bops2a <- bops2a[which(bops2a$n >= 2),]
bops2a <- bops2a %>%
group_by(customer_id) %>%
summarise(net_purchase_amount = sum(net_purchase_amount), store_number = first(store_number), gender = first(gender), age_band = first(age_band), est_income_code = first(est_income_code), ethnic_code = first(ethnic_code), bops = first(bops), numitems = first(n), homeowner_code = first(homeowner_code), length_of_residence = first(length_of_residence), child = first(child))
a = count(bops2b, customer_id)
bops2b <- merge(bops2b, a, by="customer_id", all.x=TRUE)
bops2b <- bops2b[which(bops2b$n >= 2),]
bops2b <- bops2b %>%
group_by(customer_id) %>%
summarise(net_purchase_amount = sum(net_purchase_amount), store_number = first(store_number), gender = first(gender), age_band = first(age_band), est_income_code = first(est_income_code), ethnic_code = first(ethnic_code), bops = first(bops), numitems = first(n), homeowner_code = first(homeowner_code), length_of_residence = first(length_of_residence), child = first(child))
bops2 <- rbind(bops2a, bops2b)
```
```{r echo=FALSE, warning=FALSE, message=FALSE}
bops2$missing_data <- 0
bops2$missing_data <- rowSums(is.na(bops2))
```
```{r echo=FALSE, warning=FALSE, message=FALSE}
cat("Number of na entries:", sum(is.na(bops$net_purchase_amount)))
bops2$net_purchase_amount <- log(bops2$net_purchase_amount+1)
```
```{r echo=FALSE, warning=FALSE, message=FALSE}
bops3 <- bops2[!(bops2$store_number == "5998"),]
bops3 <- within(bops3, store_number <- relevel(store_number, ref = 3))
```
```{r echo=FALSE, warning=FALSE, message=FALSE}
library(ggplot2)
bops$net_purchase_amount <- log(bops$net_purchase_amount+1)
```
```{r echo=FALSE, warning=FALSE, message=FALSE}
returns1 <- bops
returns1$summary <- as.factor(returns1$summary)
```
```{r echo=FALSE, warning=FALSE, message=FALSE}
library(aod)
library(Rcpp)
```
# 2SLS with 2 instruments (homeowner_code + length_of_residence)
```{r echo=FALSE, warning=FALSE, message=FALSE}
sls2 <- ivreg(return ~ bops * store_number + net_purchase_amount + gender + age_band + est_income_code + ethnic_code + month + summary | homeowner_code + length_of_residence + store_number + net_purchase_amount + gender + age_band + est_income_code + ethnic_code + month + summary, data = returns1)
summary(sls2,diagnostics=TRUE)
```