members <- read.csv("~/f1data/members.txt")
library(nortest)
members$city <- as.character(members$city)
members$country <- as.character(members$country)
members$bdate <- as.character(members$bdate)

members$city[members$city == ""] <- "Н/д"
members$country[members$country == ""] <- "Н/д"

city_table <- table(members$city)[order(table(members$city), decreasing = TRUE)]
par(mar = c(7, 4, 4, 2) + 0.1)
barplot(city_table[2:25], las = 2, cex.names = 0.8,
        cex.axis = 0.8, main = "Самые популярные города", ylab = "Кол-во человек")
country_table <- table(members$country)[order(table(members$country), decreasing = TRUE)]
barplot(country_table[c(1, 3:9)], las = 2, cex.names = 0.8,
        cex.axis = 0.8, main = "Самые популярные страны", ylab = "Кол-во человек")

par(mar = c(5, 7, 4, 2) + 0.1)

barplot(rev(city_table[1:25] / sum(city_table) * 100), las = 2, cex.names = 0.8,
        cex.axis = 0.8, main = "Самые популярные города", xlab = "%", horiz = TRUE)

barplot(rev(country_table[1:20] / sum(country_table) * 100), las = 2, cex.names = 0.8,
        cex.axis = 0.8, main = "Самые популярные страны", xlab = "%", horiz = TRUE)

par(mar = c(5, 4, 4, 2) + 0.1)

members$sex_text <- rep("Н/д", length(members$sex))
members$sex_text[members$sex == 1] <- "Жен."
members$sex_text[members$sex == 2] <- "Муж."
sex_table <- table(members$sex_text)
barplot(sex_table, ylab = "Кол-во человек", main = "Пол участника")
barplot(sex_table / sum(sex_table) * 100, ylab = "%", main = "Пол участника")

par(mar = c(6.5, 4, 4, 2) + 0.1)
male_name_table <- table(members$first_name[members$sex == 2])
male_name_table <- male_name_table[order(male_name_table, decreasing = TRUE)]
barplot(male_name_table[1:25], ylab = "Кол-во человек", main =
            "Самые популярные мужские имена", las = 2)

male_surname_table <- table(members$last_name[members$sex == 2])
male_surname_table <- male_surname_table[order(male_surname_table, decreasing = TRUE)]
barplot(male_surname_table[1:25], ylab = "Кол-во человек", main =
            "Самые популярные мужские фамилии", las = 2)

female_name_table <- table(members$first_name[members$sex == 1])
female_name_table <- female_name_table[order(female_name_table, decreasing = TRUE)]
barplot(female_name_table[1:25], ylab = "Кол-во человек", main =
            "Самые популярные женские имена", las = 2)


female_surname_table <- table(members$last_name[members$sex == 1])
female_surname_table <- female_surname_table[order(female_surname_table, decreasing = TRUE)]
barplot(female_surname_table[1:25], ylab = "Кол-во человек", main =
            "Самые популярные женские фамилии", las = 2)

get_year <- function(x) {
    decomp <- strsplit(x, "[.]")[[1]]
    if (length(decomp) != 3) return(NA)
    return (as.numeric(decomp[3]))
}

V_get_year <- Vectorize(get_year)

get_month <- function(x) {
    decomp <- strsplit(x, "[.]")[[1]]
    if (length(decomp) < 2) return(NA)
    return (as.numeric(decomp[2]))
}

V_get_month <- Vectorize(get_month)

members$year <- sapply(members$bdate, V_get_year)
members$month <- sapply(members$bdate, V_get_month)

valid_year <- (1:length(members$year))[members$year >= 1960]
print(pearson.test(members$year[valid_year]))
histobj <- hist(members$year[valid_year])
plot(histobj, xlab = "Год", ylab = "Кол-во участников",
     main = "Распределение года рождения участников")
multiplier <- histobj$counts / histobj$density
m <- mean(members$year[valid_year], na.rm = TRUE)
std<-sqrt(var(members$year[valid_year], na.rm = TRUE))
print(c(m, std))
curve(dnorm(x, mean=m, sd=std) * multiplier[1],
      col="darkblue", add=TRUE, yaxt="n")

valid_year <- (1:length(members$year))[members$year >= 1960 & members$sex == 2]
print(pearson.test(members$year[valid_year]))
histobj <- hist(members$year[valid_year])
plot(histobj, xlab = "Год", ylab = "Кол-во участников",
     main = "Распределение года рождения мальчиков")
multiplier <- histobj$counts / histobj$density
m <- mean(members$year[valid_year], na.rm = TRUE)
std<-sqrt(var(members$year[valid_year], na.rm = TRUE))
print(c(m, std))
curve(dnorm(x, mean=m, sd=std) * multiplier[1],
      col="darkblue", add=TRUE, yaxt="n")

valid_year <- (1:length(members$year))[members$year >= 1960 & members$sex == 1]
print(pearson.test(members$year[valid_year]))
histobj <- hist(members$year[valid_year], breaks = "Scott")
plot(histobj, xlab = "Год", ylab = "Кол-во участников",
     main = "Распределение года рождения девочек")
multiplier <- histobj$counts / histobj$density
m <- mean(members$year[valid_year], na.rm = TRUE)
std<-sqrt(var(members$year[valid_year], na.rm = TRUE))
print(c(m, std))
curve(dnorm(x, mean=m, sd=std) * multiplier[1],
      col="darkblue", add=TRUE, yaxt="n")

members$sex <- as.factor(members$sex)
fit <- aov(year ~ sex, data = members[members$year >= 1960 & members$sex != 0, ])
summary(fit)
plot(fit)

month_names <- c("Янв", "Фев", "Мар", "Апр", "Май", "Июн", "Июл", "Авг",
                 "Сен", "Окт", "Ноя", "Дек")

monthcounts <- table(members$month)
barplot(monthcounts, main = "Распределение месяца рождения участника", xlab = "Месяц",
        ylab = "Кол-во участников", names.arg = month_names)

members$sex <- as.numeric(members$sex)
msextbl <- table(members$month[members$sex > 1])
print(chisq.test(msextbl))