Wickham1.Rmd

---
title: "Ggplot basics"
output: html_document
date: "2024-07-10"
---

```{r setup, include=FALSE}
library(tidyverse)
```
#smth

```{r}
summary(mpg)

#Помимо color, можно использовать shape, alpha, size.

ggplot(data = mpg)+
  geom_point(aes(displ,hwy, color = class))

#Несколько графиков по катеориям - faset_wrap(~smth)

    ggplot(data = mpg) +
          geom_point(mapping = aes(x = displ, y = hwy)) +
          facet_wrap(~drv)
  
#linetype в smooth-Line
    ggplot(data = mpg) +
          geom_smooth(aes(displ, hwy, linetype = drv)) 
    
#Демонстрация того, что aes можно задать для всех графиков + subset, где class - одна из переменных
    ggplot(data = mpg, aes(displ, hwy)) +
      geom_point(aes(color = class)) +
      geom_smooth(
        data = filter(mpg, class == "subcompact"),
se = FALSE )
    
#Cheat sheet: https://rstudio.github.io/cheatsheets/html/data-visualization.html#geoms 
    
#Barplot 
    
 ggplot(data = diamonds) +
      geom_bar(aes(cut))
 
#geom_bar and stat_count are interchangeable 
 
  ggplot(data = diamonds) +
      stat_count(aes(cut))
  
  #Within geom_bar use fill instead of color
  
f1 <-  ggplot(data = diamonds) +
      geom_bar(mapping = aes(x = cut, color = cut))
f2 <-     ggplot(data = diamonds) +
      geom_bar(mapping = aes(x = cut, fill = cut))

#Way to unite two plots

install.packages("ggpubr")
library(ggpubr)

ggarrange(f1, f2, ncol = 2, legend = "none")

#If we want to use filling with another one variable 

    ggplot(data = diamonds) +
      geom_bar(mapping = aes(x = cut, fill = clarity))
    
#Working with the positions of the bars. "identity", "dodge" or "fill". Identity by default. Визуальный формат баров 
    
ggplot(data = diamonds, aes(cut, fill = clarity))+
geom_bar(alpha = 1/5, position = "fill")

#Визуализация может быть изменена и для скаттерплота. Jitter adds some random noise

ggplot(data = mpg) +
      geom_point(aes(displ, hwy),
        position = "jitter"
      )

#Working with coordinates. coord_flip() switches the x- and y-axes.

 ggplot(data = mpg, mapping = aes(x = class, y = hwy)) +
          geom_boxplot() +
          coord_flip()



```


```{r}

#Data Transformation with dplyr

?seq()

seq(2,20, length.out = 5)

install.packages("nycflights13")
library(nycflights13)
library(tidyverse)

flights <- nycflights13::flights

#int stands for integers. dbl stands for doubles, or real numbers. chr stands for character vectors, or strings. dttm stands for date-times (a date + a time)Y. lgl stands for logical, vectors that contain only TRUE or FALSE.fctr stands for factors, which R uses to represent categorical variables with fixed possible values.date stands for dates.

#Filter 

jan1 <- flights %>% 
  filter(day == 1, month == 1)

flights %>% 
  filter(day == 1 | month == 1)

#Alternative

filter(flights, day == 1 | month == 1)

filter(flights, !(arr_delay > 120 | dep_delay > 120))
filter(flights, arr_delay <= 120, dep_delay <= 120)

# %in% operator. Позволяет задавать условие на несколько чисел. То же самое, что "или", но меньше писать 

flights %>% 
  filter(month %in% c(11,12)) 

flights %>% 
  filter(month == 11 | month == 12) 

#Working with missing values (NAs)

data1 <- tibble(x = c(1,2, NA))

data1 %>% 
  filter(is.na(x) | x >1)

#ARRANGE (descending, ascending order)

flights %>% 
  arrange(year, month, day)

flights %>% 
  arrange(desc(arr_delay))

#SELECT

flights %>% 
  select(year, month,day)

flights %>% 
  select(year:day)

flights %>% 
  select(-(year:day))

#Select functions: starts_with("abc"), ends_with("xyz"), contains("ijk") 

#rename(new_name = old_name)

#MUTATE 

flights %>% 
  select(year:day,
 ends_with("delay"),
 distance,
 air_time) %>% 
  mutate(gain = arr_delay - dep_delay,
 speed = distance / air_time * 60)

#If I only want to keep the new columns, I can use transmute

#Summarize: na.rm = TRUE позволяет игнорировать NA (удалить их из датасета)

flights %>% 
  summarize(mean(dep_delay, na.rm = TRUE))

flights %>% 
  summarize(mean(dep_delay))

flights %>% 
  group_by(year, month, day) %>% 
  summarize(AverageDistance = mean(distance))

#n() позволяет посчитать всякое говно. Просто variable = n(). Полезно вместе с group_by

flights <- flights %>% 
  group_by(dest) %>% 
  summarise(count = n(), dist = mean(distance, na.rm = TRUE),
 delay = mean(arr_delay, na.rm = TRUE)) %>% 
  filter(count > 20, dest != "NHL")

#We can vizualize 

ggplot(data = flights, aes(dist, delay))+
  geom_point(aes(size = count), alpha = 0.25)+
  geom_smooth(se = FALSE)

#COUNTS

#комбинирование обработки данных и визуализации 

install.packages("Lahman")
library(Lahman)

batting <- as_tibble(Lahman::Batting)

batters <- batting %>%
 group_by(playerID) %>%
 summarize(
 ba = sum(H, na.rm = TRUE) / sum(AB, na.rm = TRUE),
 ab = sum(AB, na.rm = TRUE)
 )

batters %>%
 filter(ab > 100) %>%
 ggplot(mapping = aes(x = ab, y = ba)) +
 geom_point() +
 geom_smooth(se = FALSE)

#SUMMARY FUNCTIONS: mean, median, sd, IQR (interquartile range), min(), max(), first(),last(). The code is not working, there is no object with such parameters

flights %>% 
 group_by(year, month, day) %>%
 summarize(
 first_dep = first(dep_time),
 last_dep = last(dep_time, na_rm = TRUE)
 )

#COUNTS. Which destinations have the most carriers? n_distinct() - количество уникальных значений

flights %>% 
  group_by(dest) %>% 
  summarize(carriers = n_distinct(carrier)) %>%
  arrange(desc(carriers))

flights %>% 
  count(dest)

#GROUPED MUTATES!

 not_cancelled <- flights %>%
      filter(!is.na(dep_delay), !is.na(arr_delay))
 
 #Здесь можно найти минимальное и максимальное значение 
 
 not_cancelled %>%
          group_by(year, month, day) %>%
          summarize(
            first = min(dep_time),
            last = max(dep_time)
          )

  not_cancelled %>%
          group_by(year, month, day) %>%
          summarize(
            first_dep = first(dep_time),
            last_dep = last(dep_time)
          )
  
# Which destinations have the most carriers? n_distinct - позволяет определить количество уникальных входов в датасете. Count - простой подсчет 
  
   not_cancelled %>%
      group_by(dest) %>%
      summarize(carriers = n_distinct(carrier)) %>%
      arrange(desc(carriers))
   
   not_cancelled %>%
      count(dest)


betaHPD <- function(alpha,beta,p=.95,plot=FALSE,xlim=NULL,debug=FALSE){
  
  if(is.na(p) | is.nan(p) | p > 1 | p < 0)
    stop("p not between 0 and 1\n")
  
    if(alpha<=1 | beta <=1)
      stop("betaHPD only implemented for alpha and beta both > 1\n")

  ## initialize internal logical flags
  compute <- TRUE
  swap <- FALSE




```