ls()
weight
weight
lablr
labelr
nrow(labelr)
names(labelr)
names(labelr$data)
labelr$data
labelr
names(labelr)
labelr$labelr
labelr$toxic
setwd("..")
q()
n
summary(toxicity_calibrated)
qplot(labelr$toxic,type='hist')
names(labelr)
labelr$n
labelr
names(labelr)
fbyg
gghist(fbyg$weight)
hist(fbyg$weight)
hist(log(fbyg$weight))
fbyg$weight==1
all(fbyg$weight==1)
fbyg$weight[fbyg$weight != 1]
fbyg[fbyg$weight != 1]
fbyg[,fbyg$weight != 1]
fbyg[[fbyg$weight != 1]]
fbyg[fbyg$weight != 1,]
names(labelr)
summary(toxicity_calibrated)
toxicity_calibrated
val.data
names(labelr)
labelr$data
labelr
labelr[data]
labelr[data=='yg']
labelr[,data=='yg']
labelr[data=='yg',]
labelr[labelr$data=='yg']
labelr[,labelr$data=='yg']
labelr[labelr$data=='yg',]
toxicity_calibrated
summary(toxicity_calibrated)
yg3
yg3[,['toxic','toxic_pred']]
yg3 %>% select('toxic','toxic_pred')
yg3 |> select('toxic','toxic_pred')
names(yg3)
yg3[,c('toxic_pred','toxic')]
corr(yg3[,c('toxic_pred','toxic')])
cor(yg3[,c('toxic_pred','toxic')])
cor(yg3[,c('toxic_pred','toxic')],na.rm=T)
cor(yg3[,c('toxic_pred','toxic')],rm.na=T)
?cor(yg3[,c('toxic_pred','toxic')],use=
?cor
cor(yg3[,c('toxic_pred','toxic')],use='all.obs')
?cor
cor(yg3[,c('toxic_pred','toxic')],use='complete.obs')
cor(yg3[,c('toxic_pred','toxic')],use='complete.obs',method='spearman')
?predict
yg3$toxic_pred
names(preds)
preds
preds
preds$error
preds
preds
summary(errormod)
summary(errormod)
summary(preds)
names(preds)
preds
resids
qplot(resids)
resids
?predict.lm
dnorm(1)
dnorm(2)
dnorm(1)
pnorm(1)
preds
p1 + p2
p1 + p2
p1
p2
preds
preds1 <- preds
preds1$diff - preds$diff
preds1$diff 
preds1$diff - preds1$diff
preds1$diff - preds$diff
preds1$diff - preds$diff
preds1$diff - preds$diff
preds1$diff - preds$diff
preds1
preds
dnorm(-1)
dnorm(1)
pnorm(1)
pnorm(-1)
pnorm(2)
pnorm(9)
pnorm(6)
pnorm(2)
dnorm(0.95)
qnorm(0.95)
qnorm(0.841)
fulldata_preds
names(yg3)
yg3$toxic_feature_1
yg3$toxic_feature_2
yg3
yg3[,.(toxic_pred,toxic_var)]
yg3[,.(toxic_pred,toxicity_2_pred_sigma,toxicity_1_pred_sigma)]
yg3[,.(toxic_pred,toxicity_2_pred_sigma,toxicity_1_pred_sigma,cov(toxicity_2_pred,toxicity_1_pred))]
cov(1,2)
cov(c(1),c(3))
cov(c(1),c(3,2))
cov(c(1,1),c(3,2))
cov(c(1,2),c(3,2))
covterm
covterm
?cov
covterm
yg3
yg3[,.(toxic_pred,toxicity_2_pred_sigma,toxicity_1_pred_sigma,cov(toxicity_2_pred,toxicity_1_pred))]
yg3[,.(toxic_pred,toxicity_2_pred_sigma,toxicity_1_pred_sigma,toxic_var)]
yg3[,.(toxic_pred,toxicity_2_pred_sigma,toxicity_1_pred_sigma,toxic_var,toxic_sd)]
yg3
names(yg3)
print(sg)
print(sg)
1+1
library(stargazer)

stargazer(w1,w2,w3,w4,w5,t1,t2,t3,t4,t5, type="text",
          keep = c("cond1","meantox","cond1:meantox","Constant"),
          keep.stat=c("n","adj.rsq"),
          model.numbers = F,
          dep.var.labels = c("DV = Willingness to comment","DV = Toxicity of YG respondent comments"),
          covariate.labels = c("Treatment (top comments shown)",
                               "Average toxicity of top comments",
                               "Treatment $\times$ top comments toxicity",
                               "Constant"),
          add.lines = list(c("Article fixed effects","No","No","No","Yes","Yes","No","No","No","Yes","Yes")),
          star.cutoffs = c(0.05,0.01,0.005),
          notes = "Standard errors are clustered at the respondent level.",
          column.labels = c("(1)","(2)","(3)","(4)","(5)","(6)","(7)","(8)","(9)","(10)"),
          style = "apsr")

q()
n
yglabels
labelr
names(labelr)
fb
names(fb
)
fb.comment_id
fb['comment_id']
fb[,'comment_id']
labelr[,'comment_id']
names(fb)
fb.labeled
names(fb.labeled)
names(yg)
?amelia
yg
names(yg)
names(yg3)
?rbind
nrow(yg3)
nrow(yg)
yg3[,.(.N),by=.(toxic,fb)]
yg3.toimpute
names(yg3.toimpute)
yg3.toimpute
names(yg3.toimpute)
names(labelr)
nrow(yg3)
nrow(labelr)
?merge.data.table
labelr
is.data.table(labelr)
yg3.toimpute
overimp.grid
overimp.grid
?amelia
q()
n
setwd("presentations/ica_hackathon_2022/")
ls()
attach(r)
example_2_B.plot.df
library(ggplot2)
example_2_B.plot.df[(variable=='x') && (m < 1000)]
example_2_B.plot.df[(variable=='x') && (m < 1000)]
theme_set(theme_default())
theme_set(theme_minimal())
theme_set(theme_classic())
example_2_B.plot.df[(variable=='x') && (m < 1000)]
example_2_B.plot.df[(variable=='x') && (m < 1000),unique(method)]
as.factor
update.packages()
update.packages()
update.packages()
cancel
plot.df
example_2_B.plot.df
plot.df
example_2_B.plot.df
example_2_B.plot.df$method %>% unique
example_2_B.plot.df$method |> unique
example_2_B.plot.df$method  |> uniq
unique(example_2_B.plot.df$method)
example_2_B.plot.df$method
example_2_B.plot.df$method
example_2_B.plot.df$method
example_2_B.plot.df$method
example_2_B.plot.df <- r$example_2_B.plot.df
q()
n
setwd("presentations/ica_hackathon_2022/')
setwd("presentations/ica_hackathon_2022/")
example_2_B.plot.df$method
example_2_B.plot.df$method
q()
n
example_2_B.plot.df$method
example_2_B.plot.df$method
q()
n
example_2_B.plot.df$method
example_2_B.plot.df$method
q()
n
q()
n
plot.df
plot.df
plot.df[,.N,by=.(N,m)]
plot.df[,.N,by=.(N,m,method)]
plot.df[variable=='x',.N,by=.(N,m,method)]
plot.df
plot.df[(variable=='x') & (m < 1000) & (!is.na(p.true.in.ci))]
plot.df[(variable=='x') & (m != 1000) & (!is.na(p.true.in.ci))]
plot.df
?label_wrap_gen
install.packages("ggplot2")
devtools::install_github("tidyverse/ggplot2")
2
library(ggplot2)
ggplot2::version
sessioninfo()
sessionInfo()
q()
n
sessionInfo()
?scale_x_discrete
?facet_grid
plot.df
plot.df
plot.df[method="2SLS+gmm"]
plot.df[method=="2SLS+gmm"]
df <- example_2_B.plot.df
df
q()
n
plot.df
plot.df[m=50]
plot.df[m==50]
plot.df.example.2[m==50][method=2SLS+gmm]
plot.df.example.2[m==50][method==2SLS+gmm]
plot.df.example.2[(m==50) & (method==2SLS+gmm)]
plot.df.example.2[(m==50) & (method=="2SLS+gmm")]
plot.df[m==50]
plot.df.example.3
plot.df.example.3
plot.df.example.3[N=25000]
plot.df.example.3[N==25000]
plot.df
plot.df
plot.df
q()
n