mgaughan/24_deb_pkg_gov
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

first draft of mlm rdd

Browse Source
This commit is contained in:
mjgaughan 2024-04-15 09:38:41 -05:00
parent 89ffcfbe6f
commit bb1b45b348
3 changed files with 519 additions and 474 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

944
R/.Rhistory
View File

@ -1,488 +1,51 @@
pivot_longer(cols = starts_with("cnt"),
names_to = "window",
values_to = "count") |>
unnest(count))
longer <- new_test |>
pivot_longer(cols = starts_with("cnt"),
names_to = "window",
values_to = "count") |>
unnest(count)
longer
longer <- new_test |>
pivot_longer(cols = starts_with("cnt"),
names_to = "window",
values_to = "count") |>
unnest(count) |>
as.numeric(unlist(count))
longer <- new_test |>
pivot_longer(cols = starts_with("cnt"),
names_to = "window",
values_to = "count") |>
unnest(count) |>
unlist(count)
View(new_longer)
new_longer
longer <- new_test |>
pivot_longer(cols = starts_with("cnt"),
names_to = "window",
values_to = "count") |>
unnest(count) |>
unlist(count) |>
as.numeric(count)
View(new_longer)
new_longer
longer <- new_test |>
pivot_longer(cols = starts_with("cnt"),
names_to = "window",
values_to = "count") |>
unnest(count) |>
unlist(count) |>
as.numeric(count)
longer
longer <- new_test |>
pivot_longer(cols = starts_with("cnt"),
names_to = "window",
values_to = "count") |>
unnest(count) |>
unlist(count)
longer
longer <- new_test |>
pivot_longer(cols = starts_with("cnt"),
names_to = "window",
values_to = "count") |>
unnest(count)
longer
View(longer)
longer <- new_test |>
pivot_longer(cols = starts_with("cnt"),
names_to = "window",
values_to = "count") |>
unnest(count) |>
as.numeric(count)
longer
longer <- new_test |>
pivot_longer(cols = starts_with("cnt"),
names_to = "window",
values_to = "count") |>
unnest(count)
longer
library(tidyverse)
#set wd, read in data
try(setwd(dirname(rstudioapi::getActiveDocumentContext()$path)))
readme_df <- read_csv("../final_data/deb_readme_did.csv")
#preprocessing for readme_df
colnames(readme_df) <- c("upstream_vcs_link", "event_date", "event_hash", "before_all_ct", "before_mrg_ct", "after_all_ct", "after_mrg_ct", "before_auth_new", "after_commit_new", "after_auth_new", "before_commit_new")
col_order <- c("upstream_vcs_link", "event_date", "event_hash", "before_all_ct", "after_all_ct", "before_mrg_ct", "after_mrg_ct", "before_auth_new", "after_auth_new", "before_commit_new", "after_commit_new")
readme_df <- readme_df[,col_order]
readme_df$cnt_before_all <- str_split(gsub("[][]","", readme_df$before_all_cnt), ", ")
readme_df$cnt_after_all <- str_split(gsub("[][]","", readme_df$after_all_cnt), ", ")
readme_df$cnt_before_mrg <- str_split(gsub("[][]","", readme_df$before_mrg_cnt), ", ")
readme_df$cnt_after_mrg <- str_split(gsub("[][]","", readme_df$after_mrg_cnt), ", ")
drop <- c("before_all_ct", "before_mrg_ct", "after_all_ct", "after_mrg_ct")
library(tidyverse)
#set wd, read in data
try(setwd(dirname(rstudioapi::getActiveDocumentContext()$path)))
readme_df <- read_csv("../final_data/deb_readme_did.csv")
#preprocessing for readme_df
colnames(readme_df) <- c("upstream_vcs_link", "event_date", "event_hash", "before_all_ct", "before_mrg_ct", "after_all_ct", "after_mrg_ct", "before_auth_new", "after_commit_new", "after_auth_new", "before_commit_new")
col_order <- c("upstream_vcs_link", "event_date", "event_hash", "before_all_ct", "after_all_ct", "before_mrg_ct", "after_mrg_ct", "before_auth_new", "after_auth_new", "before_commit_new", "after_commit_new")
readme_df <- readme_df[,col_order]
readme_df$cnt_before_all <- str_split(gsub("[][]","", readme_df$before_all_cnt), ", ")
library(tidyverse)
#set wd, read in data
try(setwd(dirname(rstudioapi::getActiveDocumentContext()$path)))
readme_df <- read_csv("../final_data/deb_readme_did.csv")
contributing_df <- read_csv("../final_data/deb_contrib_did.csv")
#preprocessing for readme_df
colnames(readme_df) <- c("upstream_vcs_link", "event_date", "event_hash", "before_all_ct", "before_mrg_ct", "after_all_ct", "after_mrg_ct", "before_auth_new", "after_commit_new", "after_auth_new", "before_commit_new")
col_order <- c("upstream_vcs_link", "event_date", "event_hash", "before_all_ct", "after_all_ct", "before_mrg_ct", "after_mrg_ct", "before_auth_new", "after_auth_new", "before_commit_new", "after_commit_new")
readme_df <- readme_df[,col_order]
readme_df$ct_before_all <- str_split(gsub("[][]","", readme_df$before_all_ct), ", ")
readme_df$ct_after_all <- str_split(gsub("[][]","", readme_df$after_all_ct), ", ")
readme_df$ct_before_mrg <- str_split(gsub("[][]","", readme_df$before_mrg_ct), ", ")
readme_df$ct_after_mrg <- str_split(gsub("[][]","", readme_df$after_mrg_ct), ", ")
drop <- c("before_all_ct", "before_mrg_ct", "after_all_ct", "after_mrg_ct")
readme_df = readme_df[,!(names(readme_df) %in% drop)]
# as.numeric(unlist(test[1]))
# test_two <- c()
# iterator <- 0
# for (entry in test) {
# readme_df$cnt_before_all[iterator] <- as.numeric(unlist(entry))
# print(as.numeric(unlist(entry)))
# iterator <- iterator + 1
# }
# test_two
#Yes, need to expand the dataframe, but again, for the sake of clarity, do not want to until analysis step
new_test <- head(readme_df, 1)
longer <- new_test |>
pivot_longer(cols = starts_with("ct"),
names_to = "window",
values_to = "count") |>
unnest(count)
longer
View(longer)
library(tidyverse)
#set wd, read in data
try(setwd(dirname(rstudioapi::getActiveDocumentContext()$path)))
readme_df <- read_csv("../final_data/deb_readme_did.csv")
contributing_df <- read_csv("../final_data/deb_contrib_did.csv")
#preprocessing for readme_df
colnames(readme_df) <- c("upstream_vcs_link", "event_date", "event_hash", "before_all_ct", "before_mrg_ct", "after_all_ct", "after_mrg_ct", "before_auth_new", "after_commit_new", "after_auth_new", "before_commit_new")
col_order <- c("upstream_vcs_link", "event_date", "event_hash", "before_all_ct", "after_all_ct", "before_mrg_ct", "after_mrg_ct", "before_auth_new", "after_auth_new", "before_commit_new", "after_commit_new")
readme_df <- readme_df[,col_order]
readme_df$ct_before_all <- str_split(gsub("[][]","", readme_df$before_all_ct), ", ")
readme_df$ct_after_all <- str_split(gsub("[][]","", readme_df$after_all_ct), ", ")
readme_df$ct_before_mrg <- str_split(gsub("[][]","", readme_df$before_mrg_ct), ", ")
readme_df$ct_after_mrg <- str_split(gsub("[][]","", readme_df$after_mrg_ct), ", ")
drop <- c("before_all_ct", "before_mrg_ct", "after_all_ct", "after_mrg_ct")
readme_df = readme_df[,!(names(readme_df) %in% drop)]
# as.numeric(unlist(test[1]))
# test_two <- c()
# iterator <- 0
# for (entry in test) {
# readme_df$cnt_before_all[iterator] <- as.numeric(unlist(entry))
# print(as.numeric(unlist(entry)))
# iterator <- iterator + 1
# }
# test_two
#Yes, need to expand the dataframe, but again, for the sake of clarity, do not want to until analysis step
new_test <- head(readme_df, 1)
longer <- new_test |>
pivot_longer(cols = starts_with("ct"),
names_to = "window",
values_to = "count") |>
unnest(count)
longer
View(longer)
longer <- ddply(longer, "window", transform, t=seq(from=0, by=1, length.out=length(window)))
library(plyr)
longer <- ddply(longer, "window", transform, t=seq(from=0, by=1, length.out=length(window)))
View(longer)
library(plyr)
library(tidyverse)
#set wd, read in data
try(setwd(dirname(rstudioapi::getActiveDocumentContext()$path)))
readme_df <- read_csv("../final_data/deb_readme_did.csv")
contributing_df <- read_csv("../final_data/deb_contrib_did.csv")
#preprocessing for readme_df
colnames(readme_df) <- c("upstream_vcs_link", "event_date", "event_hash", "before_all_ct", "before_mrg_ct", "after_all_ct", "after_mrg_ct", "before_auth_new", "after_commit_new", "after_auth_new", "before_commit_new")
col_order <- c("upstream_vcs_link", "event_date", "event_hash", "before_all_ct", "after_all_ct", "before_mrg_ct", "after_mrg_ct", "before_auth_new", "after_auth_new", "before_commit_new", "after_commit_new")
readme_df <- readme_df[,col_order]
readme_df$ct_before_all <- str_split(gsub("[][]","", readme_df$before_all_ct), ", ")
readme_df$ct_after_all <- str_split(gsub("[][]","", readme_df$after_all_ct), ", ")
readme_df$ct_before_mrg <- str_split(gsub("[][]","", readme_df$before_mrg_ct), ", ")
readme_df$ct_after_mrg <- str_split(gsub("[][]","", readme_df$after_mrg_ct), ", ")
drop <- c("before_all_ct", "before_mrg_ct", "after_all_ct", "after_mrg_ct")
readme_df = readme_df[,!(names(readme_df) %in% drop)]
# as.numeric(unlist(test[1]))
# test_two <- c()
# iterator <- 0
# for (entry in test) {
# readme_df$cnt_before_all[iterator] <- as.numeric(unlist(entry))
# print(as.numeric(unlist(entry)))
# iterator <- iterator + 1
# }
# test_two
#Yes, need to expand the dataframe, but again, for the sake of clarity, do not want to until analysis step
new_test <- head(readme_df, 1)
longer <- new_test |>
pivot_longer(cols = starts_with("ct"),
names_to = "window",
values_to = "count") |>
unnest(count)
longer <- ddply(longer, "window", transform, t=seq(from=0, by=1, length.out=length(window)))
View(longer)
longer <- ddply(longer, strsplit("window", split="_")[-1], transform, week=seq(from=0, by=1, length.out=length(window)))
longer <- ddply(longer, strsplit(window, split="_")[-1], transform, week=seq(from=0, by=1, length.out=length(window)))
longer <- new_test |>
pivot_longer(cols = starts_with("ct"),
names_to = "window",
values_to = "count") |>
unnest(count) |>
add_column(rel = gsub("^.*_", "", window))
longer <- new_test |>
pivot_longer(cols = starts_with("ct"),
names_to = "window",
values_to = "count") |>
unnest(count) |>
add_column(rel = gsub("^.*_", "", "window"))
View(longer)
longer$rel <- gsub("^.*_", "", longer$window)
View(longer)
# as.numeric(unlist(test[1]))
# test_two <- c()
# iterator <- 0
# for (entry in test) {
# readme_df$cnt_before_all[iterator] <- as.numeric(unlist(entry))
# print(as.numeric(unlist(entry)))
# iterator <- iterator + 1
# }
# test_two
#Yes, need to expand the dataframe, but again, for the sake of clarity, do not want to until analysis step
new_test <- head(readme_df, 1)
new_testr$observation_type <- gsub("^.*_", "", new_test$window)
# as.numeric(unlist(test[1]))
# test_two <- c()
# iterator <- 0
# for (entry in test) {
# readme_df$cnt_before_all[iterator] <- as.numeric(unlist(entry))
# print(as.numeric(unlist(entry)))
# iterator <- iterator + 1
# }
# test_two
#Yes, need to expand the dataframe, but again, for the sake of clarity, do not want to until analysis step
new_test <- head(readme_df, 1)
longer <- new_test |>
pivot_longer(cols = starts_with("ct"),
names_to = "window",
values_to = "count") |>
unnest(count)
longer$observation_type <- gsub("^.*_", "", longer$window)
longer <- ddply(longer, "observation_type", transform, week=seq(from=0, by=1, length.out=length(observation_type)))
View(longer)
head(longer)
sapply(longer, class)
library(plyr)
library(tidyverse)
#set wd, read in data
try(setwd(dirname(rstudioapi::getActiveDocumentContext()$path)))
readme_df <- read_csv("../final_data/deb_readme_did.csv")
contributing_df <- read_csv("../final_data/deb_contrib_did.csv")
#preprocessing for readme_df
colnames(readme_df) <- c("upstream_vcs_link", "event_date", "event_hash", "before_all_ct", "before_mrg_ct", "after_all_ct", "after_mrg_ct", "before_auth_new", "after_commit_new", "after_auth_new", "before_commit_new")
col_order <- c("upstream_vcs_link", "event_date", "event_hash", "before_all_ct", "after_all_ct", "before_mrg_ct", "after_mrg_ct", "before_auth_new", "after_auth_new", "before_commit_new", "after_commit_new")
readme_df <- readme_df[,col_order]
readme_df$ct_before_all <- str_split(gsub("[][]","", readme_df$before_all_ct), ", ")
readme_df$ct_after_all <- str_split(gsub("[][]","", readme_df$after_all_ct), ", ")
readme_df$ct_before_mrg <- str_split(gsub("[][]","", readme_df$before_mrg_ct), ", ")
readme_df$ct_after_mrg <- str_split(gsub("[][]","", readme_df$after_mrg_ct), ", ")
drop <- c("before_all_ct", "before_mrg_ct", "after_all_ct", "after_mrg_ct")
readme_df = readme_df[,!(names(readme_df) %in% drop)]
# as.numeric(unlist(test[1]))
# test_two <- c()
# iterator <- 0
# for (entry in test) {
# readme_df$cnt_before_all[iterator] <- as.numeric(unlist(entry))
# print(as.numeric(unlist(entry)))
# iterator <- iterator + 1
# }
# test_two
#Yes, need to expand the dataframe, but again, for the sake of clarity, do not want to until analysis step
new_test <- head(readme_df, 1)
longer <- new_test |>
pivot_longer(cols = starts_with("ct"),
names_to = "window",
values_to = "count") |>
unnest(count)
longer$observation_type <- gsub("^.*_", "", longer$window)
longer <- ddply(longer, "observation_type", transform, week=seq(from=0, by=1, length.out=length(observation_type)))
View(longer)
#testing out analysis below
longer[which(longer$observation_type == all)] |>
ggplot(aes(x = week, y = count)) +
geom_point() +
geom_vline(xintercept = 26)
#testing out analysis below
longer[which(longer$observation_type == "all")] |>
ggplot(aes(x = week, y = count)) +
geom_point() +
geom_vline(xintercept = 26) geom_vline(xintercept = 26)
window <- 26
longer <- longer %>%
filter(week >= (26 - window) & week <= (26 + window))
window_num <- 26
longer <- longer %>%
filter(week >= (26 - window_num) & week <= (26 + window_num))
#testing out analysis below #testing out analysis below
longer[which(longer$observation_type == "all"),] |> longer[which(longer$observation_type == "all"),] |>
ggplot(aes(x = week, y = count)) + ggplot(aes(x = week, y = count)) +
geom_point() + geom_point() +
geom_vline(xintercept = 26) geom_vline(xintercept = 26)
View(readme_df)
# as.numeric(unlist(test[1]))
# test_two <- c()
# iterator <- 0
# for (entry in test) {
# readme_df$cnt_before_all[iterator] <- as.numeric(unlist(entry))
# print(as.numeric(unlist(entry)))
# iterator <- iterator + 1
# }
# test_two
#Yes, need to expand the dataframe, but again, for the sake of clarity, do not want to until analysis step
new_test <- readme_df[5,]
longer <- new_test |>
pivot_longer(cols = starts_with("ct"),
names_to = "window",
values_to = "count") |>
unnest(count)
longer$observation_type <- gsub("^.*_", "", longer$window)
longer <- ddply(longer, "observation_type", transform, week=seq(from=0, by=1, length.out=length(observation_type)))
#testing out analysis below
longer[which(longer$observation_type == "all"),] |>
ggplot(aes(x = week, y = count)) +
geom_point() +
geom_vline(xintercept = 26)
View(readme_df)
# as.numeric(unlist(test[1]))
# test_two <- c()
# iterator <- 0
# for (entry in test) {
# readme_df$cnt_before_all[iterator] <- as.numeric(unlist(entry))
# print(as.numeric(unlist(entry)))
# iterator <- iterator + 1
# }
# test_two
#Yes, need to expand the dataframe, but again, for the sake of clarity, do not want to until analysis step
new_test <- readme_df[76,]
longer <- new_test |>
pivot_longer(cols = starts_with("ct"),
names_to = "window",
values_to = "count") |>
unnest(count)
longer$observation_type <- gsub("^.*_", "", longer$window)
longer <- ddply(longer, "observation_type", transform, week=seq(from=0, by=1, length.out=length(observation_type)))
#testing out analysis below
longer[which(longer$observation_type == "all"),] |>
ggplot(aes(x = week, y = count)) +
geom_point() +
geom_vline(xintercept = 26)
# as.numeric(unlist(test[1]))
# test_two <- c()
# iterator <- 0
# for (entry in test) {
# readme_df$cnt_before_all[iterator] <- as.numeric(unlist(entry))
# print(as.numeric(unlist(entry)))
# iterator <- iterator + 1
# }
# test_two
#Yes, need to expand the dataframe, but again, for the sake of clarity, do not want to until analysis step
new_test <- readme_df[77,]
longer <- new_test |>
pivot_longer(cols = starts_with("ct"),
names_to = "window",
values_to = "count") |>
unnest(count)
longer$observation_type <- gsub("^.*_", "", longer$window)
longer <- ddply(longer, "observation_type", transform, week=seq(from=0, by=1, length.out=length(observation_type)))
#testing out analysis below
longer[which(longer$observation_type == "all"),] |>
ggplot(aes(x = week, y = count)) +
geom_point() +
geom_vline(xintercept = 26)
# as.numeric(unlist(test[1]))
# test_two <- c()
# iterator <- 0
# for (entry in test) {
# readme_df$cnt_before_all[iterator] <- as.numeric(unlist(entry))
# print(as.numeric(unlist(entry)))
# iterator <- iterator + 1
# }
# test_two
#Yes, need to expand the dataframe, but again, for the sake of clarity, do not want to until analysis step
new_test <- readme_df[143,]
longer <- new_test |>
pivot_longer(cols = starts_with("ct"),
names_to = "window",
values_to = "count") |>
unnest(count)
longer$observation_type <- gsub("^.*_", "", longer$window)
longer <- ddply(longer, "observation_type", transform, week=seq(from=0, by=1, length.out=length(observation_type)))
#testing out analysis below
longer[which(longer$observation_type == "all"),] |>
ggplot(aes(x = week, y = count)) +
geom_point() +
geom_vline(xintercept = 26)
# as.numeric(unlist(test[1]))
# test_two <- c()
# iterator <- 0
# for (entry in test) {
# readme_df$cnt_before_all[iterator] <- as.numeric(unlist(entry))
# print(as.numeric(unlist(entry)))
# iterator <- iterator + 1
# }
# test_two
#Yes, need to expand the dataframe, but again, for the sake of clarity, do not want to until analysis step
new_test <- readme_df[185,]
longer <- new_test |>
pivot_longer(cols = starts_with("ct"),
names_to = "window",
values_to = "count") |>
unnest(count)
longer$observation_type <- gsub("^.*_", "", longer$window)
longer <- ddply(longer, "observation_type", transform, week=seq(from=0, by=1, length.out=length(observation_type)))
#testing out analysis below
longer[which(longer$observation_type == "all"),] |>
ggplot(aes(x = week, y = count)) +
geom_point() +
geom_vline(xintercept = 26)
# as.numeric(unlist(test[1]))
# test_two <- c()
# iterator <- 0
# for (entry in test) {
# readme_df$cnt_before_all[iterator] <- as.numeric(unlist(entry))
# print(as.numeric(unlist(entry)))
# iterator <- iterator + 1
# }
# test_two
#Yes, need to expand the dataframe, but again, for the sake of clarity, do not want to until analysis step
new_test <- readme_df[231,]
longer <- new_test |>
pivot_longer(cols = starts_with("ct"),
names_to = "window",
values_to = "count") |>
unnest(count)
longer$observation_type <- gsub("^.*_", "", longer$window)
longer <- ddply(longer, "observation_type", transform, week=seq(from=0, by=1, length.out=length(observation_type)))
#testing out analysis below
longer[which(longer$observation_type == "all"),] |>
ggplot(aes(x = week, y = count)) +
geom_point() +
geom_vline(xintercept = 26)
longer[which(longer$observation_type == "all"),] |>
mutate(D = ifelse(week >= 26, 1, 0)) |>
lm(count ~ D + I(week - 26)) |>
summary()
longer[which(longer$observation_type == "all"),] |>
mutate(D = ifelse(week >= 26, 1, 0)) |>
lm(count ~ D * I(week - 26)) |>
summary()
longer[which(longer$observation_type == "all"),] |> longer[which(longer$observation_type == "all"),] |>
mutate(D = ifelse(week >= 26, 1, 0)) |> mutate(D = ifelse(week >= 26, 1, 0)) |>
lm(formula = count ~ D * I(week - 26)) |> lm(formula = count ~ D * I(week - 26)) |>
summary() summary()
longer[which(longer$observation_type == "all"),] |> longer[which(longer$observation_type == "all"),] |>
select(count, week) |> select(count, week) |>
mutate(D = ifelse(week >= 26, 1, 0)) |>
ggplot(aes(x = week, y = count, color = D)) +
geom_point() +
geom_smooth(method = "lm")
longer[which(longer$observation_type == "all"),] |>
select(count, week) |>
mutate(D = as.factor(ifelse(week >= 26, 1, 0))) |> mutate(D = as.factor(ifelse(week >= 26, 1, 0))) |>
ggplot(aes(x = week, y = count, color = D)) + ggplot(aes(x = week, y = count, color = D)) +
geom_point() + geom_point() +
geom_smooth(method = "lm") geom_smooth(se = FALSE) +
longer[which(longer$observation_type == "all"),] |> geom_vline(xintercept = 26)
select(count, week) |> window_num <- 27
mutate(D = as.factor(ifelse(week >= 26, 1, 0))) |> longer <- longer %>%
ggplot(aes(x = week, y = count, color = D)) + filter(week >= (26 - window_num) & week <= (26 + window_num))
geom_point() + # test_two <- c()
geom_smooth() # iterator <- 0
longer[which(longer$observation_type == "all"),] |> # for (entry in test) {
select(count, week) |> # readme_df$cnt_before_all[iterator] <- as.numeric(unlist(entry))
mutate(D = as.factor(ifelse(week >= 26, 1, 0))) |> # print(as.numeric(unlist(entry)))
ggplot(aes(x = week, y = count, color = D)) + # iterator <- iterator + 1
geom_point() + # }
geom_smooth() # test_two
longer[which(longer$observation_type == "all"),] |> #Yes, need to expand the dataframe, but again, for the sake of clarity, do not want to until analysis step
select(count, week) |> # https://rpubs.com/phle/r_tutorial_regression_discontinuity_design
mutate(D = as.factor(ifelse(week >= 26, 1, 0))) |> new_test <- readme_df[450,]
ggplot(aes(x = week, y = count, color = D)) + longer <- new_test |>
geom_point() + pivot_longer(cols = starts_with("ct"),
geom_smooth(aes(x = week, y = count, color = D)) names_to = "window",
longer[which(longer$observation_type == "all"),] |> values_to = "count") |>
select(count, week) |> unnest(count)
mutate(D = as.factor(ifelse(week >= 26, 1, 0))) |> longer$observation_type <- gsub("^.*_", "", longer$window)
ggplot(aes(x = week, y = count, color = D)) + longer <- ddply(longer, "observation_type", transform, week=seq(from=0, by=1, length.out=length(observation_type)))
geom_point() +
geom_smooth()
sapply(longer, class)
longer$count <- as.numeric(longer$count) longer$count <- as.numeric(longer$count)
sapply(longer, class) window_num <- 27
longer <- longer %>%
filter(week >= (26 - window_num) & week <= (26 + window_num))
#testing out analysis below #testing out analysis below
longer[which(longer$observation_type == "all"),] |> longer[which(longer$observation_type == "all"),] |>
ggplot(aes(x = week, y = count)) + ggplot(aes(x = week, y = count)) +
@ -497,16 +60,453 @@ select(count, week) |>
mutate(D = as.factor(ifelse(week >= 26, 1, 0))) |> mutate(D = as.factor(ifelse(week >= 26, 1, 0))) |>
ggplot(aes(x = week, y = count, color = D)) + ggplot(aes(x = week, y = count, color = D)) +
geom_point() + geom_point() +
geom_smooth() geom_smooth(se = FALSE) +
geom_vline(xintercept = 26)
window_num <- 20
longer <- longer %>%
filter(week >= (26 - window_num) & week <= (26 + window_num))
#testing out analysis below
longer[which(longer$observation_type == "all"),] |>
ggplot(aes(x = week, y = count)) +
geom_point() +
geom_vline(xintercept = 26)
longer[which(longer$observation_type == "all"),] |>
mutate(D = ifelse(week >= 26, 1, 0)) |>
lm(formula = count ~ D * I(week - 26)) |>
summary()
longer[which(longer$observation_type == "all"),] |> longer[which(longer$observation_type == "all"),] |>
select(count, week) |> select(count, week) |>
mutate(D = as.factor(ifelse(week >= 26, 1, 0))) |> mutate(D = as.factor(ifelse(week >= 26, 1, 0))) |>
ggplot(aes(x = week, y = count, color = D)) + ggplot(aes(x = week, y = count, color = D)) +
geom_point() + geom_point() +
geom_smooth(se = False) geom_smooth(se = FALSE) +
geom_vline(xintercept = 26)
window_num <- 4
longer <- longer %>%
filter(week >= (26 - window_num) & week <= (26 + window_num))
#testing out analysis below
longer[which(longer$observation_type == "all"),] |>
ggplot(aes(x = week, y = count)) +
geom_point() +
geom_vline(xintercept = 26)
longer[which(longer$observation_type == "all"),] |>
mutate(D = ifelse(week >= 26, 1, 0)) |>
lm(formula = count ~ D * I(week - 26)) |>
summary()
longer[which(longer$observation_type == "all"),] |> longer[which(longer$observation_type == "all"),] |>
select(count, week) |> select(count, week) |>
mutate(D = as.factor(ifelse(week >= 26, 1, 0))) |> mutate(D = as.factor(ifelse(week >= 26, 1, 0))) |>
ggplot(aes(x = week, y = count, color = D)) + ggplot(aes(x = week, y = count, color = D)) +
geom_point() + geom_point() +
geom_smooth(se = FALSE) geom_smooth(se = FALSE) +
geom_vline(xintercept = 26)
window_num <- 10
longer <- longer %>%
filter(week >= (26 - window_num) & week <= (26 + window_num))
# test_two <- c()
# iterator <- 0
# for (entry in test) {
# readme_df$cnt_before_all[iterator] <- as.numeric(unlist(entry))
# print(as.numeric(unlist(entry)))
# iterator <- iterator + 1
# }
# test_two
#Yes, need to expand the dataframe, but again, for the sake of clarity, do not want to until analysis step
# https://rpubs.com/phle/r_tutorial_regression_discontinuity_design
new_test <- readme_df[450,]
longer <- new_test |>
pivot_longer(cols = starts_with("ct"),
names_to = "window",
values_to = "count") |>
unnest(count)
longer$observation_type <- gsub("^.*_", "", longer$window)
longer <- ddply(longer, "observation_type", transform, week=seq(from=0, by=1, length.out=length(observation_type)))
longer$count <- as.numeric(longer$count)
window_num <- 10
longer <- longer %>%
filter(week >= (26 - window_num) & week <= (26 + window_num))
#testing out analysis below
longer[which(longer$observation_type == "all"),] |>
ggplot(aes(x = week, y = count)) +
geom_point() +
geom_vline(xintercept = 26)
longer[which(longer$observation_type == "all"),] |>
mutate(D = ifelse(week >= 26, 1, 0)) |>
lm(formula = count ~ D * I(week - 26)) |>
summary()
longer[which(longer$observation_type == "all"),] |>
select(count, week) |>
mutate(D = as.factor(ifelse(week >= 26, 1, 0))) |>
ggplot(aes(x = week, y = count, color = D)) +
geom_point() +
geom_smooth(se = FALSE) +
geom_vline(xintercept = 26)
# test_two <- c()
# iterator <- 0
# for (entry in test) {
# readme_df$cnt_before_all[iterator] <- as.numeric(unlist(entry))
# print(as.numeric(unlist(entry)))
# iterator <- iterator + 1
# }
# test_two
#Yes, need to expand the dataframe, but again, for the sake of clarity, do not want to until analysis step
# https://rpubs.com/phle/r_tutorial_regression_discontinuity_design
new_test <- readme_df[697,]
longer <- new_test |>
pivot_longer(cols = starts_with("ct"),
names_to = "window",
values_to = "count") |>
unnest(count)
longer$observation_type <- gsub("^.*_", "", longer$window)
longer <- ddply(longer, "observation_type", transform, week=seq(from=0, by=1, length.out=length(observation_type)))
longer$count <- as.numeric(longer$count)
window_num <- 27
longer <- longer %>%
filter(week >= (26 - window_num) & week <= (26 + window_num))
#testing out analysis below
longer[which(longer$observation_type == "all"),] |>
ggplot(aes(x = week, y = count)) +
geom_point() +
geom_vline(xintercept = 26)
longer[which(longer$observation_type == "all"),] |>
mutate(D = ifelse(week >= 26, 1, 0)) |>
lm(formula = count ~ D * I(week - 26)) |>
summary()
longer[which(longer$observation_type == "all"),] |>
select(count, week) |>
mutate(D = as.factor(ifelse(week >= 26, 1, 0))) |>
ggplot(aes(x = week, y = count, color = D)) +
geom_point() +
geom_smooth(se = FALSE) +
geom_vline(xintercept = 26)
window_num <- 13
longer <- longer %>%
filter(week >= (26 - window_num) & week <= (26 + window_num))
#testing out analysis below
longer[which(longer$observation_type == "all"),] |>
ggplot(aes(x = week, y = count)) +
geom_point() +
geom_vline(xintercept = 26)
longer[which(longer$observation_type == "all"),] |>
mutate(D = ifelse(week >= 26, 1, 0)) |>
lm(formula = count ~ D * I(week - 26)) |>
summary()
longer[which(longer$observation_type == "all"),] |>
select(count, week) |>
mutate(D = as.factor(ifelse(week >= 26, 1, 0))) |>
ggplot(aes(x = week, y = count, color = D)) +
geom_point() +
geom_smooth(se = FALSE) +
geom_vline(xintercept = 26)
longer[which(longer$observation_type == "all"),] |>
select(count, week) |>
mutate(D = as.factor(ifelse(week >= 26, 1, 0))) |>
ggplot(aes(x = week, y = count, color = D)) +
geom_point() +
geom_smooth(se = TRUE) +
geom_vline(xintercept = 26)
#testing out analysis below
longer[which(longer$observation_type == "all"),] |>
ggplot(aes(x = week, y = count)) +
geom_point() +
geom_vline(xintercept = 25.5)
longer[which(longer$observation_type == "all"),] |>
mutate(D = ifelse(week >= 26, 1, 0)) |>
lm(formula = count ~ D * I(week - 26)) |>
summary()
longer[which(longer$observation_type == "all"),] |>
select(count, week) |>
mutate(D = as.factor(ifelse(week >= 26, 1, 0))) |>
ggplot(aes(x = week, y = count, color = D)) +
geom_point() +
geom_smooth(se = TRUE) +
geom_vline(xintercept = 25.5)
#testing out analysis below
longer[which(longer$observation_type == "all"),] |>
ggplot(aes(x = week, y = count)) +
geom_point() +
geom_vline(xintercept = 26)
longer[which(longer$observation_type == "all"),] |>
mutate(D = ifelse(week >= 26, 1, 0)) |>
lm(formula = count ~ D * I(week - 26)) |>
summary()
longer[which(longer$observation_type == "all"),] |>
select(count, week) |>
mutate(D = as.factor(ifelse(week >= 26, 1, 0))) |>
ggplot(aes(x = week, y = count, color = D)) +
geom_point() +
geom_smooth(se = TRUE) +
geom_vline(xintercept = 26)
library(rdd-package)
library(rdd)
library(rdd)
# test_two <- c()
# iterator <- 0
# for (entry in test) {
# readme_df$cnt_before_all[iterator] <- as.numeric(unlist(entry))
# print(as.numeric(unlist(entry)))
# iterator <- iterator + 1
# }
# test_two
#Yes, need to expand the dataframe, but again, for the sake of clarity, do not want to until analysis step
# https://rpubs.com/phle/r_tutorial_regression_discontinuity_design
new_test <- readme_df[697,]
longer <- new_test |>
pivot_longer(cols = starts_with("ct"),
names_to = "window",
values_to = "count") |>
unnest(count)
longer$observation_type <- gsub("^.*_", "", longer$window)
longer <- ddply(longer, "observation_type", transform, week=seq(from=0, by=1, length.out=length(observation_type)))
longer$count <- as.numeric(longer$count)
#longer <- longer %>%
# filter(week >= (26 - window_num) & week <= (26 + window_num))
IKbandwidth(longer$week, longer$count, cutpoint = 26, verbose = FALSE, kernel = "triangular")
#testing out analysis below
longer[which(longer$observation_type == "all"),] |>
ggplot(aes(x = week, y = count)) +
geom_point() +
geom_vline(xintercept = 26)
longer[which(longer$observation_type == "all"),] |>
mutate(D = ifelse(week >= 26, 1, 0)) |>
lm(formula = count ~ D * I(week - 26)) |>
summary()
longer[which(longer$observation_type == "all"),] |>
select(count, week) |>
mutate(D = as.factor(ifelse(week >= 26, 1, 0))) |>
ggplot(aes(x = week, y = count, color = D)) +
geom_point() +
geom_smooth(se = TRUE) +
geom_vline(xintercept = 26)
# test_two <- c()
# iterator <- 0
# for (entry in test) {
# readme_df$cnt_before_all[iterator] <- as.numeric(unlist(entry))
# print(as.numeric(unlist(entry)))
# iterator <- iterator + 1
# }
# test_two
#Yes, need to expand the dataframe, but again, for the sake of clarity, do not want to until analysis step
# https://rpubs.com/phle/r_tutorial_regression_discontinuity_design
new_test <- readme_df[0,]
longer <- new_test |>
pivot_longer(cols = starts_with("ct"),
names_to = "window",
values_to = "count") |>
unnest(count)
longer$observation_type <- gsub("^.*_", "", longer$window)
longer <- ddply(longer, "observation_type", transform, week=seq(from=0, by=1, length.out=length(observation_type)))
longer$count <- as.numeric(longer$count)
#longer <- longer %>%
# filter(week >= (26 - window_num) & week <= (26 + window_num))
IKbandwidth(longer$week, longer$count, cutpoint = 26, verbose = FALSE, kernel = "triangular")
# test_two <- c()
# iterator <- 0
# for (entry in test) {
# readme_df$cnt_before_all[iterator] <- as.numeric(unlist(entry))
# print(as.numeric(unlist(entry)))
# iterator <- iterator + 1
# }
# test_two
#Yes, need to expand the dataframe, but again, for the sake of clarity, do not want to until analysis step
# https://rpubs.com/phle/r_tutorial_regression_discontinuity_design
new_test <- readme_df[3,]
longer <- new_test |>
pivot_longer(cols = starts_with("ct"),
names_to = "window",
values_to = "count") |>
unnest(count)
longer$observation_type <- gsub("^.*_", "", longer$window)
longer <- ddply(longer, "observation_type", transform, week=seq(from=0, by=1, length.out=length(observation_type)))
longer$count <- as.numeric(longer$count)
#longer <- longer %>%
# filter(week >= (26 - window_num) & week <= (26 + window_num))
IKbandwidth(longer$week, longer$count, cutpoint = 26, verbose = FALSE, kernel = "triangular")
# test_two <- c()
# iterator <- 0
# for (entry in test) {
# readme_df$cnt_before_all[iterator] <- as.numeric(unlist(entry))
# print(as.numeric(unlist(entry)))
# iterator <- iterator + 1
# }
# test_two
#Yes, need to expand the dataframe, but again, for the sake of clarity, do not want to until analysis step
# https://rpubs.com/phle/r_tutorial_regression_discontinuity_design
new_test <- readme_df[9,]
longer <- new_test |>
pivot_longer(cols = starts_with("ct"),
names_to = "window",
values_to = "count") |>
unnest(count)
longer$observation_type <- gsub("^.*_", "", longer$window)
longer <- ddply(longer, "observation_type", transform, week=seq(from=0, by=1, length.out=length(observation_type)))
longer$count <- as.numeric(longer$count)
#longer <- longer %>%
# filter(week >= (26 - window_num) & week <= (26 + window_num))
IKbandwidth(longer$week, longer$count, cutpoint = 26, verbose = FALSE, kernel = "triangular")
get_optimal_window <- function(project_row) {
longer <- project_row |>
pivot_longer(cols = starts_with("ct"),
names_to = "window",
values_to = "count") |>
unnest(count)
longer$observation_type <- gsub("^.*_", "", longer$window)
longer <- ddply(longer, "observation_type", transform, week=seq(from=0, by=1, length.out=length(observation_type)))
longer$count <- as.numeric(longer$count)
optimal_bandwidth <- IKbandwidth(longer$week, longer$count, cutpoint = 26, verbose = FALSE, kernel = "triangular")
return(optimal_bandwidth)
}
bandwidths <- c()
for (i in 1:nrow(readme_df)){
bandwidths <- c(bandwidths, get_optimal_window(readme_df[i,]))
}
bandwidths
mean(bandwidths)
median(bandwidths)
get_optimal_window <- function(project_row) {
longer <- project_row |>
pivot_longer(cols = starts_with("ct"),
names_to = "window",
values_to = "count") |>
unnest(count)
longer$observation_type <- gsub("^.*_", "", longer$window)
longer <- ddply(longer, "observation_type", transform, week=seq(from=0, by=1, length.out=length(observation_type)))
longer$count <- as.numeric(longer$count)
longer <- longer[which(longer$observation_type == "all"),]
optimal_bandwidth <- IKbandwidth(longer$week, longer$count, cutpoint = 26, verbose = FALSE, kernel = "triangular")
return(optimal_bandwidth)
}
bandwidths <- c()
for (i in 1:nrow(readme_df)){
bandwidths <- c(bandwidths, get_optimal_window(readme_df[i,]))
}
mean(bandwidths)
median(bandwidths)
bandwidths <- c()
for (i in 1:nrow(readme_df)){
bandwidth <- get_optimal_window(readme_df[i,])
bandwidths <- c(bandwidths, bandwidth)
}
mean(bandwidths)
median(bandwidths)
get_optimal_window <- function(project_row) {
longer <- project_row |>
pivot_longer(cols = starts_with("ct"),
names_to = "window",
values_to = "count") |>
unnest(count)
longer$observation_type <- gsub("^.*_", "", longer$window)
longer <- ddply(longer, "observation_type", transform, week=seq(from=0, by=1, length.out=length(observation_type)))
longer$count <- as.numeric(longer$count)
#this below line makes the code specific to the all-commits data
longer <- longer[which(longer$observation_type == "all"),]
result <- tryCatch({
optimal_bandwidth <- IKbandwidth(longer$week, longer$count, cutpoint = 26, verbose = FALSE, kernel = "triangular")
return(optimal_bandwidth)
}, error = function(e){
return(8)
})
}
bandwidths <- c()
for (i in 1:nrow(readme_df)){
bandwidth <- get_optimal_window(readme_df[i,])
bandwidths <- c(bandwidths, bandwidth)
}
mean(bandwidths)
median(bandwidths)
mode(bandwidths)
table(bandwidths)
mean(bandwidths) #
median(bandwidths)
# this is the file with the lmer multi-level rddAnalysis
# 0 loading the readme data in
try(setwd(dirname(rstudioapi::getActiveDocumentContext()$path)))
readme_df <- read_csv("../final_data/deb_readme_did.csv")
# 1 preprocessing
colnames(readme_df) <- c("upstream_vcs_link", "event_date", "event_hash", "before_all_ct", "before_mrg_ct", "after_all_ct", "after_mrg_ct", "before_auth_new", "after_commit_new", "after_auth_new", "before_commit_new")
col_order <- c("upstream_vcs_link", "event_date", "event_hash", "before_all_ct", "after_all_ct", "before_mrg_ct", "after_mrg_ct", "before_auth_new", "after_auth_new", "before_commit_new", "after_commit_new")
readme_df <- readme_df[,col_order]
readme_df$ct_before_all <- str_split(gsub("[][]","", readme_df$before_all_ct), ", ")
readme_df$ct_after_all <- str_split(gsub("[][]","", readme_df$after_all_ct), ", ")
readme_df$ct_before_mrg <- str_split(gsub("[][]","", readme_df$before_mrg_ct), ", ")
readme_df$ct_after_mrg <- str_split(gsub("[][]","", readme_df$after_mrg_ct), ", ")
drop <- c("before_all_ct", "before_mrg_ct", "after_all_ct", "after_mrg_ct")
readme_df = readme_df[,!(names(readme_df) %in% drop)]
# 2 some expansion needs to happens for each project
expand_timeseries <- function(project_row) {
longer <- project_row |>
pivot_longer(cols = starts_with("ct"),
names_to = "window",
values_to = "count") |>
unnest(count)
longer$observation_type <- gsub("^.*_", "", longer$window)
longer <- ddply(longer, "observation_type", transform, week=seq(from=0, by=1, length.out=length(observation_type)))
longer$count <- as.numeric(longer$count)
#longer <- longer[which(longer$observation_type == "all"),]
return(longer)
}
expanded_data <- expand_timeseries(readme_df[1,])
for (i in 2:nrow(readme_df)){
expanded_data <- rbind(expanded_data, expand_timeseries(readme_df[i,]))
}
View(expanded_data)
View(expanded_data)
View(expanded_data)
View(expanded_data)
View(expanded_data)
get_optimal_window <- function(project_row) {
longer <- project_row |>
pivot_longer(cols = starts_with("ct"),
names_to = "window",
values_to = "count") |>
unnest(count)
longer$observation_type <- gsub("^.*_", "", longer$window)
longer <- ddply(longer, "observation_type", transform, week=seq(from=0, by=1, length.out=length(observation_type)))
longer$count <- as.numeric(longer$count)
#this below line makes the code specific to the all-commits data
longer <- longer[which(longer$observation_type == "all"),]
result <- tryCatch({
#Imbens-Kalyanaraman Optimal Bandwidth Calculation
optimal_bandwidth <- IKbandwidth(longer$week, longer$count, cutpoint = 26, verbose = FALSE, kernel = "triangular")
return(optimal_bandwidth)
}, error = function(e){
return(9)
})
}
#this just gets the optimal bandwith window for each project and then appends to lists
bandwidths <- c()
for (i in 1:nrow(readme_df)){
bandwidth <- get_optimal_window(readme_df[i,])
bandwidths <- c(bandwidths, bandwidth)
}
mean(bandwidths) #8.574233
median(bandwidths) #8.363088
table(bandwidths)
#filter out the timewindows
window_num <- 8
expanded_data |>
filter(week >= (26 - window_num) & week <= (26 + window_num))
expanded_data |>
filter(week >= (26 - window_num) & week <= (26 + window_num))
# 3 rdd in lmer analysis
library(lme4)
draft_model <- lmer(count ~ D * I(week - 26) + upstream_vcs_link, data=expanded_data[which(longer$observation_type == "all"),])
expanded_data |>
filter(week >= (26 - window_num) & week <= (26 + window_num)) |>
mutate(D = ifelse(week >= 26, 1, 0))
# 3 rdd in lmer analysis
library(lme4)
draft_model <- lmer(count ~ D * I(week - 26) + upstream_vcs_link, data=expanded_data[which(longer$observation_type == "all"),])
summary(draft_model)
View(expanded_data)
#filter out the timewindows
window_num <- 8
expanded_data <- expanded_data |>
filter(week >= (26 - window_num) & week <= (26 + window_num)) |>
mutate(D = ifelse(week >= 26, 1, 0))
draft_model <- lmer(count ~ D * I(week - 26) + upstream_vcs_link, data=expanded_data[which(longer$observation_type == "all"),])
summary(draft_model)
draft_model <- lmer(count ~ D * I(week - 26) + upstream_vcs_link, REML=FALSE, data=expanded_data[which(longer$observation_type == "all"),])
draft_model <- lmer(count ~ D * I(week - 26) + upstream_vcs_link, REML=FALSE, data=expanded_data[which(longer$observation_type == "all"),])
draft_model <- lmer(count ~ D * I(week - 26) + (1|upstream_vcs_link), REML=FALSE, data=expanded_data[which(longer$observation_type == "all"),])
summary(draft_model)

40
R/didAnalysis.R
View File

@ -0,0 +1,40 @@
# this is the file with the lmer multi-level rddAnalysis
# 0 loading the readme data in
try(setwd(dirname(rstudioapi::getActiveDocumentContext()$path)))
readme_df <- read_csv("../final_data/deb_readme_did.csv")
# 1 preprocessing
colnames(readme_df) <- c("upstream_vcs_link", "event_date", "event_hash", "before_all_ct", "before_mrg_ct", "after_all_ct", "after_mrg_ct", "before_auth_new", "after_commit_new", "after_auth_new", "before_commit_new")
col_order <- c("upstream_vcs_link", "event_date", "event_hash", "before_all_ct", "after_all_ct", "before_mrg_ct", "after_mrg_ct", "before_auth_new", "after_auth_new", "before_commit_new", "after_commit_new")
readme_df <- readme_df[,col_order]
readme_df$ct_before_all <- str_split(gsub("[][]","", readme_df$before_all_ct), ", ")
readme_df$ct_after_all <- str_split(gsub("[][]","", readme_df$after_all_ct), ", ")
readme_df$ct_before_mrg <- str_split(gsub("[][]","", readme_df$before_mrg_ct), ", ")
readme_df$ct_after_mrg <- str_split(gsub("[][]","", readme_df$after_mrg_ct), ", ")
drop <- c("before_all_ct", "before_mrg_ct", "after_all_ct", "after_mrg_ct")
readme_df = readme_df[,!(names(readme_df) %in% drop)]
# 2 some expansion needs to happens for each project
expand_timeseries <- function(project_row) {
longer <- project_row |>
pivot_longer(cols = starts_with("ct"),
names_to = "window",
values_to = "count") |>
unnest(count)
longer$observation_type <- gsub("^.*_", "", longer$window)
longer <- ddply(longer, "observation_type", transform, week=seq(from=0, by=1, length.out=length(observation_type)))
longer$count <- as.numeric(longer$count)
#longer <- longer[which(longer$observation_type == "all"),]
return(longer)
}
expanded_data <- expand_timeseries(readme_df[1,])
for (i in 2:nrow(readme_df)){
expanded_data <- rbind(expanded_data, expand_timeseries(readme_df[i,]))
}
#filter out the timewindows
window_num <- 8
expanded_data <- expanded_data |>
filter(week >= (26 - window_num) & week <= (26 + window_num)) |>
mutate(D = ifelse(week >= 26, 1, 0))
# 3 rdd in lmer analysis
library(lme4)
draft_model <- lmer(count ~ D * I(week - 26) + (1|upstream_vcs_link), REML=FALSE, data=expanded_data[which(longer$observation_type == "all"),])
summary(draft_model)

9
R/didCleaning.R
View File

@ -52,19 +52,24 @@ get_optimal_window <- function(project_row) {
optimal_bandwidth <- IKbandwidth(longer$week, longer$count, cutpoint = 26, verbose = FALSE, kernel = "triangular") optimal_bandwidth <- IKbandwidth(longer$week, longer$count, cutpoint = 26, verbose = FALSE, kernel = "triangular")
return(optimal_bandwidth) return(optimal_bandwidth)
}, error = function(e){ }, error = function(e){
#have tested it with multiple different error-values and all medians/means still hover around 8
return(8) return(8)
}) })
} }
#this just gets the optimal bandwith window for each project and then appends to lists
bandwidths <- c() bandwidths <- c()
for (i in 1:nrow(readme_df)){ for (i in 1:nrow(readme_df)){
bandwidth <- get_optimal_window(readme_df[i,]) bandwidth <- get_optimal_window(readme_df[i,])
bandwidths <- c(bandwidths, bandwidth) bandwidths <- c(bandwidths, bandwidth)
} }
mean(bandwidths) #8.574233 mean(bandwidths)
#8.574233
median(bandwidths) #8.363088 median(bandwidths) #8.363088
table(bandwidths) table(bandwidths)
#window_num <- 13 #from this, I think setting the bandwidth to 8 weeks, two months, the floor
# of both the median and mean calculations
#longer <- longer %>% #longer <- longer %>%
# filter(week >= (26 - window_num) & week <= (26 + window_num)) # filter(week >= (26 - window_num) & week <= (26 + window_num))