library(ggplot2)
library(scales)
library(dplyr)

theme_set(theme_gray(base_size=22))
REGRESSION_METHOD <- "loess"
SIZE <- 3

## df <- read.csv('./unam.csv', header=TRUE)
df <- read.csv('./full.csv', header=TRUE)

Latin_America <- c('Argentina', 'Belize', 'Bolivia', 'Brazil', 'Chile',
                   'Colombia', 'Costa Rica', 'Cuba', 'Dominican Republic',
                   'Ecuador', 'El Salvador', 'French Guiana', 'Guatemala',
                   'Guyana', 'Haiti', 'Honduras', 'Mexico', 'Nicaragua',
                   'Panama', 'Paraguay', 'Peru',
                   'Puerto Rico', 'Suriname', 'Uruguay', 'Venezuela')
Iberic_World <- c(Latin_America, 'Spain', 'Portugal', 'Mozambique',
                  'Equatorial Guinea', 'Angola', 'Cape Verde',
                  'São Tomé and Príncipe')
North_America <- c('Belize','Costa Rica', 'Cuba', 'Dominican Republic',
                   'El Salvador', 'Guatemala',
                   'Haiti', 'Honduras', 'Mexico', 'Nicaragua',
                   'Panama', 'Puerto Rico', 'United States', 'Canada',
                   'Jamaica', 'Bahamas')

df$Latin_America <- sapply(df$Country, function(x) x %in% Latin_America)
df$North_America <- sapply(df$Country, function(x) x %in% North_America)
df$Iberic_World <- sapply(df$Country, function(x) x %in% Iberic_World)
df$Mexico <- sapply(df$Country, function(x) x == "Mexico")

## Reorder levels
df$Region <- factor(df$Region,
                    levels=c("World", "Latin America", "Mexico", "Iberic World", "North America"))
df$Ranking <- factor(df$Ranking,
                     levels=c("QS", "SCImago", "CSIC Spain", "URAP", "NTU", "ARWU", "THE"))

## Censoring:
##
## Outlier
## df <- df[df$Ranking != "THE", ]
## Missing year in many sources
df <- df[df$Year != 2025, ]
## Very incomplete , biases regression
## df <- df[! (df$Region == "Latin America" & df$Ranking == "QS"), ]

integer_breaks <- function(n) {
    function(x) pretty_breaks(n)(x) %>% .[. %% 1 == 0]
}

svg("./Ranking_UNAM_historical_THE.2024.svg", width=16, height=10)

ggplot(data=df, mapping=aes(x=Year, y=Rank)) +
    geom_smooth(method=REGRESSION_METHOD, size=SIZE, color="gray", se=TRUE) +
    geom_jitter(mapping=aes(x=Year, y=Rank, color=Ranking, shape=Ranking),
                height=0,
                width=.1,
                data=df,
                size=SIZE,
                alpha=.5) +
    geom_line(mapping=aes(x=Year, y=Rank, color=Ranking), data=df) +
    scale_y_continuous(breaks = integer_breaks(n=6), trans = "reverse") +
    scale_x_continuous(breaks = integer_breaks(n=6)) +
    theme_minimal() + theme_minimal(base_size = 22) +
    theme(legend.position = c(.85, .2), axis.text.x = element_text(angle = 45, hjust = 1)) +
    ggtitle("Academic rankings UNAM 2012-2024") +
    facet_wrap(~Region, scales = "free_y")

dev.off()