MATH2349 Semester 1, 2019

# This is a chunk where you can load the necessary packages required to reproduce the report. Here are some example packages, you may add others if you require  

library(readr)
library(tidyr)
library(dplyr)
library(Hmisc)
library(outliers)

# This is an R chunk for reading the WHO data. Provide your R codes here:
who <- read_csv("WHO.csv")

Parsed with column specification:
cols(
  .default = col_double(),
  country = col_character(),
  iso2 = col_character(),
  iso3 = col_character()
)
See spec(...) for full column specifications.

# This is an R chunk for tidy task 1. Provide your R codes here:

who1 <- who %>% gather(code, value, 5:60)
who1

NA

# This is an R chunk for tidy task 2. Provide your R codes here:

who2 <- who1 %>% separate(code, c("new", "var", "sex"), sep = "_")
who2 <- who2 %>% separate(sex, c("sex", "age"), sep = "(?<=[A-Za-z])(?=[0-9999])")
who2

NA

# This is an R chunk for tidy task 3. Provide your R codes here:
who3 <- who2 %>% spread(var, value)
who3

NA
NA
NA

# This is a chunk for Task 4. Provide your R codes here: 
who4 <- who3 %>% mutate(age = factor(who3$age, levels = c("014", "1524", "2534", "3544", "4554", "5564", "65"),
                  labels = c("<15", "15-24", "25-34", "35-44", "45-54", "55-64", "65>="), ordered = TRUE))
who4

NA

# This is a chunk for Task 5. Provide your R codes here: 
who5 <- who4 %>% select(-c(iso2, new))
who5

WHO_subset <- who5 %>% filter(country == c("Albania", "Japan", "Greece"))

longer object length is not a multiple of shorter object length

WHO_subset

NA
NA

# This is an R chunk for reading the Species and Surveys data sets. Provide your R codes here:
species <- read_csv("species.csv")

Parsed with column specification:
cols(
  species_id = col_character(),
  genus = col_character(),
  species = col_character(),
  taxa = col_character()
)

surveys <- read_csv("surveys.csv")

Parsed with column specification:
cols(
  record_id = col_double(),
  month = col_double(),
  day = col_double(),
  year = col_double(),
  species_id = col_character(),
  sex = col_character(),
  hindfoot_length = col_double(),
  weight = col_double()
)

# This is a chunk for Task 6. Provide your R codes here: 
surveys_combined <- left_join(surveys, species, key = "species_id")

Joining, by = "species_id"

surveys_combined

NA

# This is a chunk for Task 7. Provide your R codes here: 
surveys_combined %>% filter(species_id == "DM") %>% group_by(month) %>% 
  summarise('Avg Weight' = mean(weight, na.rm = TRUE), 'Avg Foot Lenght' = mean(hindfoot_length, na.rm = TRUE))

NA

# This is a chunk for Task 8. Provide your R codes here: 
surveys_combined_year <- surveys_combined %>% filter(year == "2000")
surveys_combined_year

#Displaying the number of NA observations in weight for the year 2000
surveys_combined_year %>% group_by(species_id) %>% summarise('Weight NAs' = sum(is.na(weight)))  

#Imputing the means of each species weight into NA observations
surveys_weight_imputed <- surveys_combined_year %>% group_by(species_id) %>% mutate(weight = ifelse(is.na(weight), mean(weight, na.rm = TRUE), weight))
surveys_weight_imputed %>% group_by(species_id) %>% summarise('Weight NAs' = sum(is.na(weight)))

NA
NA
NA
NA

# This is a chunk for Task 9. Provide your R codes here: 

#Checking weight column for infinite values
sum(is.infinite(surveys_weight_imputed$weight))

[1] 0

#Checking weight column for NaN values
sum(is.nan(surveys_weight_imputed$weight))

[1] 68

#No infinite values were found, however the weight column has 68 NaN values.

#This is due to some species not having any weight values, so they were unable to be imputed using the mean weight of that species. The code run in task 8 would have imputed the mean of NA into species without any weight observations resulting in NaN.

# This is a chunk for Task 10. Provide your R codes here: 
#Creating a boxplot of the data
surveys_combined$hindfoot_length %>% boxplot(Main = "Boxplot of Hindfoot Length") 

#From the boxplot, 2 outliers can be observed

#Calculating z-scores for data
zscores <- na.omit(surveys_combined$hindfoot_length) %>% scores(type = "z")
#Displaying z-scores above 3 as that is the limit for outliers
length (which( abs(zscores) >3))

[1] 4

#Following the z-score method, there are only 4 outliers in a dataset of 35549 observations

#To manage the outliers, we will simply exlclude them since they are a small percentage of the dataset
hindfoot_new <- surveys_combined$hindfoot_length[ -which(abs (zscores)>3)]
length(surveys_combined$hindfoot_length)

[1] 35549

length(hindfoot_new)

[1] 35545

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