diff --git a/Julie/DataCall.R b/Julie/DataCall.R
new file mode 100644
index 0000000..d771ffc
--- /dev/null
+++ b/Julie/DataCall.R
@@ -0,0 +1,36 @@
+library(tidyverse)
+library(gridExtra)
+library(Hmisc)
+library(reldist)
+
+# nipa_income
+
+nipa <- read_csv("data-hackaton/nipa_income.csv", col_names = TRUE, )
+
+
+# Sample A
+
+sampA <- read_csv("data-hackaton/sample_A.csv", 
+                  col_names = TRUE, 
+                  col_types = cols(
+                    new_hrs = col_double(),
+                    new_wage = col_double()
+                  ))
+
+# Sample B 
+
+sampB <- read_csv("data-hackaton/sample_B.csv", 
+                  col_names = TRUE, 
+                  col_types = cols(
+                    new_hrs = col_double(),
+                    new_wage = col_double()
+                  ))
+
+# Sample C
+
+sampC <- read_csv("data-hackaton/sample_C.csv", 
+                  col_names = TRUE, 
+                  col_types = cols(
+                    new_hrs = col_double(),
+                    new_wage = col_double()
+                  ))
\ No newline at end of file
diff --git a/Julie/Figure 9.R b/Julie/Figure 9.R
new file mode 100644
index 0000000..931c005
--- /dev/null
+++ b/Julie/Figure 9.R	
@@ -0,0 +1,26 @@
+quant = sampC %>% select(year, hh_earnings_equiv, hh_wgt)
+
+sampgender = function(data){
+
+  wage1 = data %>% 
+    filter(!is.na(hh_earnings_equiv)) %>%
+    group_by(year) %>%
+    summarise(wage = mean(hh_earnings_equiv)) %>%
+  
+  
+  B = wage2 %>% summarise(bias = mean(bias))
+  b = B$bias
+  
+  final1 = wage1 %>%
+    mutate(wage1 = wage - b) %>%
+    select(year, wage1)
+  
+  final2 = wage2 %>%
+    select(year, new_wage) %>%
+    filter(year>1975)
+  
+  data = bind_rows(final1, final2) %>% 
+    mutate(wage = ifelse(year<1976, wage1, new_wage))
+  
+  return(data)
+}
diff --git a/Julie/Figure1.R b/Julie/Figure1.R
new file mode 100644
index 0000000..21b91dd
--- /dev/null
+++ b/Julie/Figure1.R
@@ -0,0 +1,35 @@
+inc = sampA %>% select(year, labor_income, pre_tax_income, cpi, wgt) %>%
+  mutate(true_year = year - 1, 
+         real_income = labor_income/cpi * 100, 
+         real_pre_tax = pre_tax_income/cpi * 100) %>%
+  group_by(true_year) %>%
+  summarise(real_income = log(weighted.mean(real_income, wgt)), 
+            real_pre_tax = log(weighted.mean(real_pre_tax, wgt)), 
+            cpi = mean(cpi)) %>%
+  mutate(type = "CPS") %>%
+  select(true_year, real_income, real_pre_tax, type)
+
+nipa = nipa %>% 
+  mutate(real_income = log(salary_nipa), 
+         real_pre_tax = log(total_income_nipa), 
+         type = "NIPA") %>%
+  select(true_year, real_income, real_pre_tax, type)
+
+g1 = ggplot(data = nipa, aes(x = true_year, y = real_income)) + 
+  geom_point(color = "red") + 
+  geom_line(data = inc, color = "blue") + 
+  theme_bw() + 
+  labs(title = "Labor Income Per Capita", 
+       x = "Year", 
+       y = "")
+
+g2 = ggplot(data = nipa, aes(x = true_year, y = real_pre_tax)) + 
+  geom_point(color = "red") + 
+  geom_line(data = inc, color = "blue", ) + 
+  theme_bw() + 
+  labs(title = "Pre-Tax Income Per Capita", 
+       x = "Year", 
+       y = "")
+
+grid.arrange(g1, g2, ncol = 2)
+    
diff --git a/Julie/Figure2.R b/Julie/Figure2.R
new file mode 100644
index 0000000..84fe880
--- /dev/null
+++ b/Julie/Figure2.R
@@ -0,0 +1,115 @@
+sampgender = function(g){
+
+data2 = sampA %>% filter(between(age, 25, 60)) %>% 
+  select(sex, year, wage, new_wage)
+
+wage1 = data2 %>% 
+  select(year, wage, new_wage, sex) %>%
+  filter(!is.na(wage), year < 1976) %>%
+  filter(sex == g) %>%
+  group_by(year) %>%
+  summarise(wage = mean(wage)) %>%
+  mutate(new_wage = NA) 
+
+wage2 = data2 %>% 
+  select(year, wage, new_wage, sex) %>%
+  filter(!is.na(wage), year > 1975) %>%
+  filter(!is.na(new_wage)) %>%
+  filter(sex == g) %>%
+  group_by(year) %>%
+  summarise(wage = mean(wage), 
+            new_wage = mean(new_wage)) %>%
+  mutate(bias = wage - new_wage)
+
+B = wage2 %>% summarise(bias = mean(bias))
+b = B$bias
+
+final1 = wage1 %>%
+  mutate(wage1 = wage - b) %>%
+  select(year, wage1)
+
+final2 = wage2 %>%
+  select(year, new_wage) %>%
+  filter(year>1975)
+
+data = bind_rows(final1, final2) %>% 
+  mutate(wage = ifelse(year<1976, wage1, new_wage))
+
+return(data)
+}
+
+wagem = sampgender(1)
+wagef = sampgender(2)
+
+wm = ggplot(wagem, aes(x = year, y = wage)) + geom_line() + 
+  labs(title = "Average Male Wage", 
+       x = "Year", 
+       y = "")
+
+wf = ggplot(wagef, aes(x = year, y = wage)) + geom_line() + 
+  labs(title = "Average Female Wage", 
+       x = "Year", 
+       y = "")
+
+sampgender2 = function(g){
+  
+  data2 = sampA %>% filter(between(age, 25, 60)) %>% 
+    select(sex, year, hrs, new_hrs)
+  
+  hrs1 = data2 %>% 
+    select(year, hrs, new_hrs, sex) %>%
+    filter(!is.na(hrs), year < 1976) %>%
+    filter(sex == g) %>%
+    group_by(year) %>%
+    summarise(hrs = mean(hrs)) %>%
+    mutate(new_hrs = NA) 
+  
+  hrs2 = data2 %>% 
+    select(year, hrs, new_hrs, sex) %>%
+    filter(!is.na(hrs), year > 1975) %>%
+    filter(!is.na(new_hrs)) %>%
+    filter(sex == g) %>%
+    group_by(year) %>%
+    summarise(hrs = mean(hrs), 
+              new_hrs = mean(new_hrs)) %>%
+    mutate(bias = hrs - new_hrs)
+  
+  B = hrs2 %>% summarise(bias = mean(bias))
+  b = B$bias
+  
+  final1 = hrs1 %>%
+    mutate(hrs1 = hrs - b) %>%
+    select(year, hrs1)
+  
+  final2 = hrs2 %>%
+    select(year, new_hrs) %>%
+    filter(year>1975)
+  
+  data = bind_rows(final1, final2) %>% 
+    mutate(hrs = ifelse(year<1976, hrs1, new_hrs))
+  
+  return(data)
+}
+
+hours1 = sampgender2(1)
+hours2 = sampgender2(2)
+
+hm = ggplot(hours1, aes(x = year, y = hrs)) + geom_line() + 
+  labs(title = "Average Male Annual hours", 
+       x = "Year", 
+       y = "")
+
+hf = ggplot(hours2, aes(x = year, y = hrs)) + geom_line() + 
+  labs(title = "Average Female Annual hours", 
+       x = "Year", 
+       y = "")
+
+grid.arrange(wm, wf, hm, hf, ncol = 2)
+
+
+
+
+
+
+
+  
\ No newline at end of file
diff --git a/Julie/Figure4.R b/Julie/Figure4.R
new file mode 100644
index 0000000..4bfaedc
--- /dev/null
+++ b/Julie/Figure4.R
@@ -0,0 +1,163 @@
+data3 = sampA %>% select(year, wage, new_wage, sex, wgt)
+
+variance = function(data, g){
+  
+    wage1 = data %>% 
+    filter(!is.na(wage), year < 1976) %>%
+    filter(sex == g) %>%
+    group_by(year) %>%
+    mutate(log = log(wage)) %>%
+    dplyr::summarise(wage = wtd.var(log, wgt)) %>%
+    mutate(new_wage = NA) 
+  
+  wage2 = data %>% 
+    filter(!is.na(wage), year > 1975) %>%
+    filter(!is.na(new_wage)) %>%
+    filter(sex == g) %>%
+    group_by(year) %>%
+    mutate(log = log(wage), 
+           logn = log(new_wage)) %>%
+    dplyr::summarise(wage = wtd.var(log, wgt), 
+              new_wage = wtd.var(logn, wgt)) %>%
+    mutate(bias = wage - new_wage)
+  
+  B = wage2 %>% dplyr::summarise(bias = mean(bias))
+  b = B$bias
+  
+  final1 = wage1 %>%
+    mutate(wage1 = wage - b) %>%
+    select(year, wage1)
+  
+  final2 = wage2 %>%
+    select(year, new_wage) %>%
+    filter(year>1975)
+  
+  data = bind_rows(final1, final2) %>% 
+    mutate(wage = ifelse(year<1976, wage1, new_wage))
+  
+  return(data)
+}
+
+var1 = variance(data3, 1) %>% mutate(gndr = "Male")
+var2 = variance(data3, 2) %>% mutate(gndr = "Female")
+var = bind_rows(var1, var2)
+
+gvar = ggplot(var, aes(x = year, y = wage)) + 
+  geom_line(aes(linetype = gndr, colour = gndr)) + 
+  labs(title = "Variance of log hourly wages",
+       x = "Year",
+       y = "")
+
+quant = function(data, g, q, t){
+  
+  wage1 = data %>% 
+    filter(!is.na(wage), year < 1976) %>%
+    filter(sex == g) %>%
+    group_by(year) %>%
+    dplyr::summarise(wage50 = wtd.quantile(wage, probs = c(q)), 
+              wage10 = wtd.quantile(wage, probs = c(t))) %>%
+    mutate(wage = wage50/wage10, 
+           w_wage = NA) 
+  
+  wage2 = data %>% 
+    filter(!is.na(wage), year > 1975) %>%
+    filter(!is.na(new_wage)) %>%
+    filter(sex == g) %>%
+    group_by(year) %>%
+    dplyr::summarise(wage50 = wtd.quantile(wage, wgt, probs = c(q)), 
+              wage10 = wtd.quantile(wage, wgt, probs = c(t)), 
+              nwage50 = wtd.quantile(new_wage, wgt, probs = c(q)), 
+              nwage10 = wtd.quantile(new_wage, wgt, probs = c(t))) %>%
+    mutate(wage = wage50/wage10,
+           nwage = wage50/wage10, 
+           bias = wage-nwage)
+  
+  B = wage2 %>% dplyr::summarise(bias = mean(bias))
+  b = B$bias
+  
+  final1 = wage1 %>%
+    mutate(wage1 = wage - b) %>%
+    select(year, wage1)
+  
+  final2 = wage2 %>%
+    select(year, nwage) %>%
+    filter(year>1975)
+  
+  data = bind_rows(final1, final2) %>% 
+    mutate(wage = ifelse(year<1976, wage1, nwage))
+  
+  return(data)
+}
+
+quant1 = quant(data3, 1, 0.5, 0.1) %>% mutate(gndr = "Male")
+quant2 = quant(data3, 2 ,0.5, 0.1) %>% mutate(gndr = "Female")
+quant = bind_rows(quant1, quant2)
+
+gquant1 = ggplot(quant, aes(x = year, y = wage)) + 
+  geom_line(aes(linetype = gndr, colour = gndr)) + 
+  labs(title = "P50 - P10 ratio of hourly wages",
+       x = "Year",
+       y = "")
+
+quant11 = quant(data3, 1, 0.9, 0.5) %>% mutate(gndr = "Male")
+quant21 = quant(data3, 2 ,0.9, 0.5) %>% mutate(gndr = "Female")
+QUant = bind_rows(quant11, quant21)
+
+gquant2 = ggplot(QUant, aes(x = year, y = wage)) + 
+  geom_line(aes(linetype = gndr, colour = gndr)) + 
+  labs(title = "P90 - 510 ratio of hourly wages",
+       x = "Year",
+       y = "")
+
+
+gini = function(data, g){
+  
+  wage1 = data %>% 
+    filter(!is.na(wage), year < 1976) %>%
+    filter(sex == g) %>%
+    group_by(year) %>%
+    dplyr::summarise(wage = reldist::gini(wage, wgt)) %>%
+    mutate(new_wage = NA) 
+  
+  wage2 = data %>% 
+    filter(!is.na(wage), year > 1975) %>%
+    filter(!is.na(new_wage)) %>%
+    filter(sex == g) %>%
+    group_by(year) %>%
+    dplyr::summarise(wage = reldist::gini(wage, wgt), 
+              new_wage = reldist::gini(new_wage, wgt)) %>%
+    mutate(bias = wage - new_wage)
+  
+  B = wage2 %>% dplyr::summarise(bias = mean(bias))
+  b = B$bias
+  
+  final1 = wage1 %>%
+    mutate(wage1 = wage - b) %>%
+    select(year, wage1)
+  
+  final2 = wage2 %>%
+    select(year, new_wage) %>%
+    filter(year>1975)
+  
+  data = bind_rows(final1, final2) %>% 
+    mutate(wage = ifelse(year<1976, wage1, new_wage))
+  
+  return(data)
+}
+
+gini1 = gini(data3, 1) %>% mutate(gndr = "Male")
+gini2 = gini(data3, 2) %>% mutate(gndr = "Female")
+gini = bind_rows(gini1, gini2)
+
+gini = ggplot(gini, aes(x = year, y = wage)) + 
+  geom_line(aes(linetype = gndr, colour = gndr)) + 
+  labs(title = "Gini Coefficient of hourly wages",
+       x = "Year",
+       y = "")
+
+grid.arrange(gvar, gini,gquant1, gquant2, ncol = 2)
+
+
+
+
+
diff --git a/Julie/Table2.R b/Julie/Table2.R
new file mode 100644
index 0000000..a89b9fa
--- /dev/null
+++ b/Julie/Table2.R
@@ -0,0 +1,53 @@
+
+sampB = sampB %>% mutate(yind = "PROB")
+
+
+
+sampB$yind = ifelse(between(sampB$year, 1968, 1980), "1967-1979", 
+                               ifelse(between(sampB$year, 1981, 1990), "1980-1989", 
+                               ifelse(between(sampB$year, 1991, 2000), "1990-1999", 
+                               ifelse(sampB$year > 2000, "from 2000", sampB$yind))))
+
+age = sampB %>% group_by(yind, sex) %>%
+  summarise(av = weighted.mean(age,wgt)) %>%
+  spread(key = yind, value = av) %>%
+  mutate(name = c("Avg male age", "Avg female age")) %>%
+  select(-sex)
+
+edu16m = sampB %>% select(yind, edu, sex) %>%
+  filter(sex == 1 & edu >= 16) %>%
+  group_by(yind) %>%
+  summarise(n = n()) %>%
+  mutate(eff = c(nrow(filter(sampB, yind == "1967-1979" & sex == 1)), 
+                 nrow(filter(sampB, yind == "1980-1989" & sex == 1)), 
+                 nrow(filter(sampB, yind == "1990-1999" & sex == 1)), 
+                 nrow(filter(sampB, yind == "from 2000" & sex == 1)))) %>%
+  mutate(edu16 = n * 100 / eff) %>%
+  select(yind, edu16) %>%
+  spread(key = yind, value = edu16) %>%
+  mutate(name = c("% male > 16 years edu"))
+
+edu16f = sampB %>% select(yind, edu, sex) %>%
+  filter(sex == 2 & edu >= 16) %>%
+  group_by(yind) %>%
+  summarise(n = n()) %>%
+  mutate(eff = c(nrow(filter(sampB, yind == "1967-1979" & sex == 2)), 
+                 nrow(filter(sampB, yind == "1980-1989" & sex == 2)), 
+                 nrow(filter(sampB, yind == "1990-1999" & sex == 2)), 
+                 nrow(filter(sampB, yind == "from 2000" & sex == 2)))) %>%
+  mutate(edu16 = n * 100 / eff) %>%
+  select(yind, edu16) %>%
+  spread(key = yind, value = edu16) %>%
+  mutate(name = c("% female > 16 years edu"))
+
+table2 = bind_rows(age, edu16m, edu16f) %>% select(name,"1967-1979", "1980-1989", 
+                                                   "1990-1999", "from 2000")
+table2
+
+
+
+
+
+
+
+