{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "# 0.3.2 - Introduction to Data in R - Part 2\n",
        "\n",
        "COMET Team <br> *Colby Grimes, Colby Chamber, Jonathan Graves, Manas\n",
        "Mridul, Valeria Zolla*  \n",
        "2023-07-09\n",
        "\n",
        "## Outline\n",
        "\n",
        "### Prerequisites\n",
        "\n",
        "-   Introduction to Data in R - Part 1\n",
        "\n",
        "### Learning Outcomes\n",
        "\n",
        "In this notebook, you will learn about:\n",
        "\n",
        "-   Filtering, segmenting, combining and cleaning datasets with\n",
        "    `tidyverse` and base R\n",
        "-   Reshaping and presenting data in different kinds of formats\n",
        "\n",
        "### References\n",
        "\n",
        "-   [Introduction to Probability and Statistics Using\n",
        "    R](https://mran.microsoft.com/snapshot/2018-09-28/web/packages/IPSUR/vignettes/IPSUR.pdf)\n",
        "-   [DSCI 100 Textbook](https://datasciencebook.ca/index.html)"
      ],
      "id": "e4d82a6a-90bf-431e-9326-638b65fd8b30"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "# run this cell\n",
        "source(\"getting_started_intro_to_data_tests_2.r\")\n",
        "# loading in our packages\n",
        "library(tidyverse)\n",
        "library(haven)"
      ],
      "id": "1339157e-6691-4f4b-a469-26e67a6e8a18"
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "## Part 1: Data Cleaning\n",
        "\n",
        "The World Development Indicators (WDI) is the primary World Bank\n",
        "collection of development indicators, compiled from officially\n",
        "recognized international sources.\n",
        "\n",
        "We have used World Bank’s DataBank applet to select and import some\n",
        "macro and development-related time series data for the countries Canada,\n",
        "India, Mexico, South Africa, and Zimbabwe for years 2014-2022."
      ],
      "id": "e6c0ced2-5e12-4ade-8e3f-5851e0de08e8"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "# importing required packages\n",
        "library(readr)\n",
        "library(dplyr)\n",
        "library(tidyr)\n",
        "library(stringr)\n",
        "\n",
        "#Loading the CSV file using the URL \n",
        "file_path <- \"datasets/WDI_raw.csv\"\n",
        "\n",
        "wdi <- read_csv(file_path) %>%\n",
        "       rename(Series_Code = \"Series Code\", Series_Name = \"Series Name\",\n",
        "       Country = \"Country Name\", Country_Code = \"Country Code\")\n",
        "\n",
        "head(wdi, 10)"
      ],
      "id": "9947aad7-de0a-40ae-97f5-1cdc3023e6bc"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "dim(wdi)"
      ],
      "id": "4342fe21-1645-441c-a3fe-7740b9d42f24"
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "The data frame is in long format. Each unique value in `Series_Name` an\n",
        "entry for a row, and not a column. As an example:\n",
        "\n",
        "| Country | Year |   Var | Value |\n",
        "|--------:|-----:|------:|------:|\n",
        "|  Canada | 2017 | Var A | Value |\n",
        "|  Canada | 2017 | Var B | Value |\n",
        "|  Canada | 2018 | Var C | Value |\n",
        "|  Canada | 2018 | Var B | Value |\n",
        "|  Canada | 2018 | Var C | Value |\n",
        "\n",
        "A simpler version of the data frame in a **wide-format** could look like\n",
        "this:\n",
        "\n",
        "| Country | Year | Var A | Var B | Var C |\n",
        "|--------:|-----:|------:|------:|------:|\n",
        "|  Canada | 2017 | Value | Value | Value |\n",
        "|  Canada | 2018 | Value | Value | Value |\n",
        "\n",
        "> **Note**: look up the functions `pivot_wider()` and `pivot_longer()`\n",
        "> to learn how to switch between long and wide formats.\n",
        "\n",
        "While `Series_Name` contains descriptions for each of the series in the\n",
        "data frame, `Series_Code` offers a handy way to group variables.\n",
        "\n",
        "Our `Series_Code` variable follow a taxonomy system. For example, any\n",
        "code starting with `AG` belongs to a *family of series* related to the\n",
        "state of agriculture in that country. Let’s see the unique series\n",
        "families and their sub series names."
      ],
      "id": "5f6b4ee8-9ac1-4dea-96aa-851449ed6b8d"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "Series_Families <- wdi %>%\n",
        "                   mutate(Series_Family = substr(`Series_Code`, 1, 2)) %>%    # substring first two characters\n",
        "                   group_by(Series_Family) %>%    # group by unique vals\n",
        "                   summarize(Series_Names_Under_Family = paste(unique(`Series_Name`), collapse = \", \")) # for each family, find and paste all `Series_Name`\n",
        "\n",
        "head(Series_Families, 13)"
      ],
      "id": "0a4904f5-f291-4d0b-84e4-9ba9082c8a19"
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "Let’s create a data frame with series families related to access to\n",
        "institutions. Those are grouped under `SE` (Schooling), `FX` (Financial\n",
        "Institutions) and `EG` (Electricity). Let’s call our dataset\n",
        "`access_wdi`."
      ],
      "id": "db145c9c-78b1-4a58-8161-17c81135352b"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "prefixes <- c(\"SE\", \"EG\", \"FX\")\n",
        "\n",
        "access_wdi <- wdi %>% \n",
        "              filter(str_detect(`Series_Code`, str_c(\"^\", prefixes, collapse = \"|\")))\n",
        "\n",
        "access_wdi <- access_wdi %>%  \n",
        "              rename_with(~str_extract(., \"\\\\d{4}\"), starts_with(\"20\")) # rename year columns to follow the 2XYZ format.\n",
        "\n",
        "head(access_wdi, 5)"
      ],
      "id": "0d54e6f0-0b0b-4705-86a8-386bf5c7e1c4"
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "### Handling missing values\n",
        "\n",
        "We are interested in seeing if there are any specific years for when the\n",
        "data was not collected or is missing for *most* of the variables/series.\n",
        "\n",
        "Let’s look at each of the year columns and count the number of rows with\n",
        "`NA`."
      ],
      "id": "ee9414cf-c8cb-469f-bea3-d8a5636ef48a"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "year_columns <- c(\"2018\", \"2019\", \"2020\", \"2021\", \"2022\")\n",
        "\n",
        "missing_counts <- sapply(year_columns, function(column) sum(is.na(access_wdi[[column]]))) # create custom function to count missing values\n",
        "\n",
        "print(missing_counts / nrow(access_wdi))"
      ],
      "id": "a3921033-be3c-42fd-bfce-171a938bc710"
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "Looks like 98% of the data is missing for the year 2022. It probably\n",
        "makes sense to drop the 2022 column altogether."
      ],
      "id": "ce40eaa2-c606-4233-ab78-817073bd6a33"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "access_wdi <- access_wdi %>% \n",
        "              select(-`2022`)\n",
        "\n",
        "head(access_wdi, 20)"
      ],
      "id": "93f08b09-81c6-4605-8056-63964dff7db4"
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "Now let’s check for missing values *row-wise* - whether there are lots\n",
        "of NAs for a country-series combination. Let’s check if there are any\n",
        "particular country-series combination for which the data is missing for\n",
        "more than 2 years."
      ],
      "id": "77539016-f569-4cf2-864c-840a09870c5c"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "# create new column with NA count by columns for each row\n",
        "access_wdi$count_na <- rowSums(is.na(access_wdi))\n",
        "\n",
        "# sort by number of missing vals\n",
        "access_wdi <- access_wdi %>% \n",
        "              arrange(desc(count_na))\n",
        "\n",
        "select(access_wdi, c(\"Country_Code\", \"Series_Code\", \"count_na\"))"
      ],
      "id": "3b1da73d-0416-4d13-a3d8-12bf3a071d3a"
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "This data frame shows that we don’t have *any* data for series beginning\n",
        "with the `SE` (Schooling) prefix.\n",
        "\n",
        "> **Think Deeper**: how could you be systematic when choosing whether to\n",
        "> drop or not NA values? Our panel datasets is indexed by country and\n",
        "> series, and we’re interested in yearly values over time. Think about\n",
        "> the options we have to drop NAs: (1) Dropping a series altogether (2)\n",
        "> Dropping specific countries (3) Dropping specific rows (i.e.,\n",
        "> country-series pairs)…\n",
        "\n",
        "Let’s create an array with `Series_Code` to be dropped from our dataset."
      ],
      "id": "8b6f2970-7cdf-4a8c-b158-72cb5a325cf5"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "# creating an array of `Series_Code` that need to be dropped from access_wdi\n",
        "\n",
        "to_drop <- access_wdi %>%\n",
        "           filter(count_na > 2) %>%\n",
        "           distinct(Series_Code) %>% unique() \n",
        "\n",
        "to_drop <- unlist(to_drop, use.names = FALSE)\n",
        "\n",
        "to_drop"
      ],
      "id": "feefdfb7-9d43-46fd-891d-67f660641244"
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "> **Think Deeper**: what if dropping the missing values leads to low\n",
        "> sample size issues? One approach is to import data from alternative\n",
        "> sources and *fill them into* our original data set. This topic is\n",
        "> beyond the scope of this notebook, but think about the implications of\n",
        "> filling in data to the study.\n",
        "\n",
        "Let’s now drop all the rows where `Series_Code` matches any of the codes\n",
        "in `to_drop`, and save the resulting data frame as a new version of\n",
        "`access_wdi`."
      ],
      "id": "e045647d-559f-46cc-92ad-ca45f320636e"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "filtered_access_wdi <- access_wdi %>%\n",
        "                       filter(!(Series_Code %in% to_drop))\n",
        "\n",
        "filtered_access_wdi"
      ],
      "id": "e0dd2934-0a97-4461-83fb-fdf66aebe1be"
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "Now the only variables left in this data frame are the `EG` variables,\n",
        "which indicate the levels of access to electricity and other power\n",
        "sources within the countries.\n",
        "\n",
        "This dataset is clearly not appropriate to answer question about the\n",
        "access to the overall access to institutions; however, it could be\n",
        "extremely useful if the scope of the research is narrowed to access to\n",
        "power utilities. For example, we could use the dataset to visualize the\n",
        "growth in access to energy across the countries over the last 5 years.\n",
        "\n",
        "### Test your knowledge\n",
        "\n",
        "Before you complete the following exercises, take a moment to review the\n",
        "section above. Are there any functions you’re not familiar with? Now is\n",
        "a good time to learn them.\n",
        "\n",
        "> **Note**: by running `?function_name` you can get a detailed\n",
        "> description of any of the functions used in this notebook.\n",
        "\n",
        "What are some possible ways to deal with NA values? Select all that\n",
        "apply\n",
        "\n",
        "1.  Drop them row-wise\n",
        "2.  Drop them column-wise\n",
        "3.  Fill them in\n",
        "4.  Change the scope of the research"
      ],
      "id": "3c9c06d1-5c06-49be-b684-9dce69a38752"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "# Enter your answer as a string (in alphabetical order) with your answer choices ex: \"ACD\"\n",
        "\n",
        "answer_1 <- \n",
        "\n",
        "test_1()"
      ],
      "id": "4d5b1bc1-5913-41aa-9651-11c4f377724e"
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "What is normally the best way to deal with missing data?\n",
        "\n",
        "1.  Drop them row-wise\n",
        "2.  Drop them column-wise\n",
        "3.  Fill them in\n",
        "4.  Change the scope of the research\n",
        "5.  Impossible to tell"
      ],
      "id": "a46b542c-68a5-4417-ad69-0ec6b8482396"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "# Enter your answer as either \"A\", \"B\", \"C\", \"D\", or \"E\"\n",
        "\n",
        "answer_2 <- \n",
        "\n",
        "test_2()"
      ],
      "id": "777b5046-1860-420b-ae93-14f454363fb3"
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "## Part 2: Merging data frames in R\n",
        "\n",
        "Now let’s take a step back and consider an example of merging data\n",
        "frames. Our WDI data set has macro information on *national incomes*,\n",
        "*CABs*, *Bank Capital to Assets Ratios,* and various kinds of *CPIA*\n",
        "ratings. Let’s extract that data and merge it with data from the\n",
        "Quarterly Public Debt (QPD) data base. The QPD is exactly what you think\n",
        "it is: a record of sovereign debt managed by the World Bank and the\n",
        "International Monetary Fund (IMF).\n",
        "\n",
        "First, to our `wdi` dataset."
      ],
      "id": "05532ecc-4f84-4c9c-a7e3-75e8758eac15"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "# identify the relevant series families \n",
        "prefixes <- c(\"NY\", \"FD\", \"FB\", \"IQ\", \"BN\")\n",
        "\n",
        "# subset `wdi` and create `macro_wdi`\n",
        "macro_wdi <- wdi %>% \n",
        "             filter(str_detect(`Series_Code`, str_c(\"^\", prefixes, collapse = \"|\"))) %>% \n",
        "             rename_with(~str_extract(., \"\\\\d{4}\"), starts_with(\"20\"))\n",
        "  \n",
        "macro_wdi"
      ],
      "id": "86588752-e101-45fd-95c9-8b8a07407bdb"
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "Again, take note of the missing values! It’s sad that we’re missing the\n",
        "CPIA rating variables (starting with `IQ`) for all the countries\n",
        "(Canada, India, Mexico, South Africa and Indonesia). Let’s drop the CPIA\n",
        "rows."
      ],
      "id": "bb0112eb-74e5-43c0-87a3-fae75653cdeb"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "macro_wdi <- macro_wdi %>% \n",
        "             filter(!(Series_Code %in% c('IQ.CPA.FINS.XQ', 'IQ.CPA.FISP.XQ', 'IQ.CPA.MACR.XQ', 'IQ.CPA.PROP.XQ', 'IQ.CPA.PROT.XQ', 'IQ.CPA.DEBT.XQ', 'IQ.CPA.TRAD.XQ')))\n",
        "\n",
        "c(macro_wdi$Series_Code %>% \n",
        "unique(), macro_wdi$Series_Name %>% \n",
        "unique())"
      ],
      "id": "2807059e-bef7-4e34-b04b-6559e4eed41b"
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "So these are the variables we’ll include from the WDI data frame! Now,\n",
        "let’s load the QPD dataset."
      ],
      "id": "bbb42d37-b85f-428e-8dba-60a26ec3a297"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "file_path_2 <- \"datasets/qpd.csv\"\n",
        "\n",
        "qpd <- read_csv(file_path_2) %>%\n",
        "       rename(Series_Code = \"Series Code\", Series_Name = \"Series Name\",\n",
        "              Country = \"Country Name\", Country_Code = \"Country Code\")\n",
        "\n",
        "head(qpd, 25)"
      ],
      "id": "6d1a075e-5e52-4a33-bc04-190fe7d02ec3"
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "The series data in QPD is stored on a quarter-by-year basis. We can\n",
        "aggregate the data *column-wise* to get yearly amounts.\n",
        "\n",
        "> **Note**: R usually throws an error if you’re telling it to sum over\n",
        "> certain rows/columns that include NA values. We resolve this by\n",
        "> setting the parameter `na.rm = TRUE`, which literally means “NA remove\n",
        "> is TRUE”.\n",
        "\n",
        "Before we aggregate the data, let’s check the number of periods for\n",
        "which data is missing. Again, we’ll look at the country-series\n",
        "combinations. We do that below with a loop, telling R to *manually* go\n",
        "over each unique row, count the number of NAs along the period columns,\n",
        "and then store the result in another dataframe called `status`."
      ],
      "id": "fcbd1ce2-f6ef-41b9-9991-cae8eac95ca9"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "status <- data.frame()    # empty data-frame that is placeholder for the final data\n",
        "Series_Codes <- qpd$Series_Code %>% unique()    # gets all `Series_Codes` to iterate over\n",
        "Countries <- qpd$Country_Code %>% unique()     # gets all `Country_Codes` to iterate over \n",
        "\n",
        "for (country_code in Countries) {\n",
        "    select <- filter(qpd, Country_Code == country_code)    # first filter for the indexed country with `country_code`\n",
        "  \n",
        "    for (series_code in Series_Codes) {\n",
        "        select_further <- filter(select, Series_Code == series_code)    # then filter by for the indexed series with `series_code`\n",
        "        cols_to_check <- select(select_further, c(\"2018Q1 [YR2018Q1]\", \"2018Q2 [YR2018Q2]\", \"2018Q3 [YR2018Q3]\", \"2018Q4 [YR2018Q4]\",    # then select the period columns \n",
        "                                              \"2019Q1 [YR2019Q1]\", \"2019Q2 [YR2019Q2]\", \"2019Q3 [YR2019Q3]\", \"2019Q4 [YR2019Q4]\",\n",
        "                                              \"2020Q1 [YR2020Q1]\", \"2020Q2 [YR2020Q2]\", \"2020Q3 [YR2020Q3]\", \"2020Q4 [YR2020Q4]\",\n",
        "                                              \"2021Q1 [YR2021Q1]\", \"2021Q2 [YR2021Q2]\", \"2021Q3 [YR2021Q3]\", \"2021Q4 [YR2021Q4]\",\n",
        "                                              \"2022Q1 [YR2022Q1]\", \"2022Q2 [YR2022Q2]\", \"2022Q3 [YR2022Q3]\", \"2022Q4 [YR2022Q4]\"))\n",
        "        \n",
        "        # the result will be a single row of period columns for each unique `Country_Code`, `Series_Code` combination\n",
        "        \n",
        "        na_count <- sum(is.na(cols_to_check))    # finally, store the value of NAs\n",
        "        result <- data.frame(Country_Code = country_code, Series_Code = series_code, na_count = na_count)\n",
        "        status <- rbind(status, result)    # appends the result to the status dataframe and iterate over\n",
        "  }\n",
        "}\n",
        "\n",
        "head(status)"
      ],
      "id": "79c318e7-d509-4df4-be9a-70b5c4cb3a3e"
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "As expected, `status` has three columns: `Country_Code`, `Series_Code`,\n",
        "and the `na_count`. Now let’s check which combinations have data missing\n",
        "for less than 20 periods."
      ],
      "id": "91c40748-5555-47e8-a103-8de0482bf27c"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "status_to_drop <- status %>% filter(na_count < 20 & na_count > 0) # strictly less than 20\n",
        "status_to_drop"
      ],
      "id": "8c110964-ab92-4e0d-87e4-b65249ff2014"
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "> **Note**: we filter by `(na_count < 20 & na_count > 0)` because if the\n",
        "> data is missing for *all* of the 20 columns, the aggregate function\n",
        "> will take care of the missing values and return `0` as the yearly\n",
        "> aggregate value. However, if data is missing for strictly less than 20\n",
        "> period columns, the yearly aggregate values will be under-estimated.\n",
        "\n",
        "These are the countries-series pairs which we must drop from the\n",
        "data-frame.\n",
        "\n",
        "> **Note**: by storing our exploration’s results in `status`, we can\n",
        "> re-visit our decision to drop values anytime we want. Such proper\n",
        "> documentation builds trust around the validity of the aggregate\n",
        "> computations!\n",
        "\n",
        "Let’s now use `anti_join()` to drop any rows from `qpd` that match the\n",
        "`Country_Code`, `Series_Code` pairs in `status_to_drop`."
      ],
      "id": "c2bd0c0f-6ea8-4119-9622-885309e42471"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "qpd_filtered <- anti_join(qpd, status_to_drop, by = c(\"Country_Code\", \"Series_Code\"))\n",
        "qpd_filtered"
      ],
      "id": "d361b1b6-a346-4236-b01b-fe787204ca04"
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "> **Note**: `anti_join()` is a function that removes rows from a\n",
        "> dataframe that have matching values in specified columns with another\n",
        "> dataframe. In this context, it is used to drop rows from qpd that\n",
        "> match the `Country_Code` and `Series_Code` pairs in `status_to_drop`,\n",
        "> resulting in the filtered dataframe `qpd_filtered`.\n",
        "\n",
        "The code below tells R how to manually go over each unique combination\n",
        "of `Country`, `Series_Code` values and aggregate quaterly values by\n",
        "year. To learn exactly what each line does, head to the Appendix!"
      ],
      "id": "5e3c5c46-5108-4f0d-b365-d17305835f94"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "# pivot the data from wide to long format, creating separate rows for each quarterly value\n",
        "qpd_long <- qpd_filtered %>%\n",
        "  tidyr::pivot_longer(starts_with(\"20\"), names_to = \"quarter\", values_to = \"value\")\n",
        "\n",
        "# extract the year from the \"quarter\" column\n",
        "qpd_extracted <- qpd_long %>%\n",
        "  dplyr::mutate(year = stringr::str_extract(quarter, \"\\\\d+\"))\n",
        "\n",
        "# group the data by country, series, and year for aggregation\n",
        "qpd_grouped <- qpd_extracted %>%\n",
        "  dplyr::group_by(Country_Code, Country, Series_Code, Series_Name, year)\n",
        "\n",
        "# calculate the sum of values for each combination of country, series, and year\n",
        "qpd_summarized <- qpd_grouped %>%\n",
        "  dplyr::summarise(total_value = sum(value, na.rm = TRUE))\n",
        "\n",
        "# pivot the data back to wide format, with separate columns for each year\n",
        "qpd_aggregated <- qpd_summarized %>%\n",
        "  tidyr::pivot_wider(names_from = year, values_from = total_value, names_prefix = \"year_\")\n",
        "\n",
        "qpd_aggregated"
      ],
      "id": "f6d38fac-9631-431e-af08-c57baec189fe"
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "Notice the zeroes in the data frame? As explained earlier, these occur\n",
        "due to certain pairs of missing data for all of the 20 time periods. We\n",
        "could filter out these observations if we had to!\n",
        "\n",
        "### Performing the merge\n",
        "\n",
        "Now let’s create a new data frame which includes both the macro\n",
        "variables from WDI and the QPD data. We’ll use `rbind()` to append\n",
        "observations from `qpd_aggregated` to `macro_wdi`."
      ],
      "id": "993d4929-136f-4520-a35b-242a411dba7f"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "# since the yearly column names are different, we can rename them to match\n",
        "colnames(qpd_aggregated) <- colnames(macro_wdi)\n",
        "\n",
        "# combine the data frames using rbind\n",
        "df_macro <- rbind(macro_wdi, qpd_aggregated)\n",
        "\n",
        "# print the dimensions of the combined data frame\n",
        "print(dim(df_macro))\n",
        "\n",
        "# view the combined data frame\n",
        "df_macro"
      ],
      "id": "1c1e86a5-74bc-4744-8ade-db7a247033c4"
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "## Part 3: Other Types of Merges\n",
        "\n",
        "Let’s look at a **horizontal merge**. Suppose we were to update\n",
        "`df_macro` the following values for the year 2023."
      ],
      "id": "e3883f76-aea0-4190-9269-fbfa1c6738ec"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "values_2023 <- c(-46898267372, -25575220073, 1457067267, 81641557999, -59663613794, 79677936994, 88935053721, 32159558497, 25822808780, -87642745906, -58805085020, -64688649494, 37404569332, -23179256357, 53968284000, -460151583, 43523701653, 98381218966, -23992964113, 55489044264, 86941046221, -57571495743, 30334753217, -74888980808, -46555866254, -22777181491, -97321933368, -23522408586, 73938169144, -31930200662, -3583976906, 19913165085)\n",
        "\n",
        "Country <- c(\"India\",  \"India\", \"Canada\", \"Canada\", \"Honduras\", \"Honduras\", \"Indonesia\", \"Indonesia\", \"South Africa\", \"South Africa\", \"Mexico\",  \"Mexico\", \"United Kingdom\", \"United Kingdom\", \"United States\", \"United States\", \"China\", \"China\", \"Hong Kong SAR, China\", \"Hong Kong SAR, China\", \"Netherlands\", \"Netherlands\", \"Egypt, Arab Rep.\", \"Egypt, Arab Rep.\", \"Georgia\", \"Georgia\", \"Slovak Republic\", \"Slovak Republic\", \"Georgia\", \"Honduras\", \"Indonesia\", \"Slovak Republic\")\n",
        "\n",
        "Series_Code <- rep(df_macro$Series_Code, length.out = length(values_2023))\n",
        "Series_Name <- rep(df_macro$Series_Name, length.out = length(values_2023))\n",
        "\n",
        "series_2023 <- data.frame( Country = Country, Series_Code = Series_Code, Series_Name = Series_Name, values_2023 = values_2023)\n",
        "\n",
        "head(series_2023, 5)"
      ],
      "id": "d71f0451-d97e-40ff-bd39-4bef444f57da"
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "We can use `merge()` to add the 2023 values column into `df_macro`,\n",
        "specifiying that we merge by `Country`, `Series_Code`, and\n",
        "`Series_Name`."
      ],
      "id": "fb7269b4-ead0-4f64-9024-537f099b625f"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "df_macro_2 <- merge(df_macro, series_2023, by = c(\"Country\", \"Series_Code\", \"Series_Name\"))\n",
        "\n",
        "df_macro_2 <- df_macro_2 %>% \n",
        "              rename(\"2023\" = \"values_2023\")\n",
        "\n",
        "head(df_macro_2, 5)"
      ],
      "id": "e85ae86b-9a05-4c88-8cf8-58efc119642e"
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "Take a look at the image below to learn about other methods for merging.\n",
        "\n",
        "<figure>\n",
        "<img src=\"attachment:media/join_visual.png\"\n",
        "alt=\"Merging Methods - COMET Team\" />\n",
        "<figcaption aria-hidden=\"true\">Merging Methods - COMET Team</figcaption>\n",
        "</figure>\n",
        "\n",
        "As illustrated in the image, a `full_join()` and `right_join()` are\n",
        "great for merging data sources in situations when we are particularly\n",
        "interested in the issue of missing matches.For simpler cases,\n",
        "`inner_join()` is ideal when you only want to include fully matched\n",
        "observations in the final data set.\n",
        "\n",
        "### Test your knowledge\n",
        "\n",
        "Debug the code below to calculate the mean of the vector `set`. Store\n",
        "the answer in `answer_3`."
      ],
      "attachments": {
        "media/join_visual.png": {
          "image/png": "iVBORw0KGgoAAAANSUhEUgAABdkAAAR4CAYAAAD65t9sAAAACXBIWXMAAAsTAAALEwEAmpwYAAAA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\nTkSuQmCC\n"
        }
      },
      "id": "f48334c6-06f7-4f9c-b109-ea7862f07f04"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "set"
      ],
      "id": "af74b9a8-4f82-4ba5-8997-bfe92382bd5b"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "mean(set)"
      ],
      "id": "812970b9-1214-45e0-a3a4-897c7b80ee69"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "answer_3 <- \n",
        "\n",
        "test_3()"
      ],
      "id": "c2e7fc0c-de4a-4e43-b0df-ca5750ed1639"
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "Suppose you have two data frames, `df1` and `df2`. What line of code\n",
        "should you run if you want to perform a merge and preserve the entries\n",
        "of `df1`?\n",
        "\n",
        "1.  `anti_join(df1, df2)`\n",
        "2.  `right_join(df1, df2)`\n",
        "3.  `inner_join(df1, df2)`\n",
        "4.  `left_join(df1, df2)`\n",
        "5.  Impossible to tell without knowing the entries"
      ],
      "id": "65bc6ffe-5163-4bed-b2e0-27ba206094e3"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "# Enter your answer as either \"A\", \"B\", \"C\", \"D\", or \"E\"\n",
        "\n",
        "answer_4 <- \n",
        "\n",
        "test_4()"
      ],
      "id": "be481839-79c7-4ca7-acc6-20d3cc4a46d4"
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "## Appendix\n",
        "\n",
        "### More on the wrangling code that aggregates period values in QPD\n",
        "\n",
        "The following code should produce 5 different data-frames that\n",
        "incrementally show how our wrangling code for calculating the yearly\n",
        "aggregates for each Country and Series."
      ],
      "id": "5706e6fb-bc0d-4eb6-821f-6189668a5624"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "# select a specific country and series for demonstration (e.g., \"USA\" and \"GDP\")\n",
        "country_code <- \"ALB\"\n",
        "series_code <- \"DP.DOD.DSCD.CR.PS.CD\"\n",
        "\n",
        "# filter the data for the specific country and series\n",
        "qpd_example <- qpd_filtered %>%\n",
        "  dplyr::filter(Country_Code == country_code, Series_Code == series_code)\n",
        "\n",
        "# pivot the data from wide to long format, creating separate rows for each quarterly value\n",
        "qpd_long <- qpd_example %>%\n",
        "  tidyr::pivot_longer(starts_with(\"20\"), names_to = \"quarter\", values_to = \"value\")\n",
        "\n",
        "# show the intermediate result: qpd_long\n",
        "qpd_long\n",
        "\n",
        "# extract the year from the \"quarter\" column\n",
        "qpd_extracted <- qpd_long %>%\n",
        "  dplyr::mutate(year = stringr::str_extract(quarter, \"\\\\d+\"))\n",
        "\n",
        "# show the intermediate result: qpd_extracted\n",
        "qpd_extracted\n",
        "\n",
        "# group the data by country, series, and year for aggregation\n",
        "qpd_grouped <- qpd_extracted %>%\n",
        "  dplyr::group_by(Country_Code, Country, Series_Code, Series_Name, year)\n",
        "\n",
        "# show the intermediate result: qpd_grouped\n",
        "qpd_grouped\n",
        "\n",
        "# calculate the sum of values for each combination of country, series, and year\n",
        "qpd_summarized <- qpd_grouped %>%\n",
        "  dplyr::summarise(total_value = sum(value, na.rm = TRUE))\n",
        "\n",
        "# show the intermediate result: qpd_summarized\n",
        "qpd_summarized\n",
        "\n",
        "# pivot the data back to wide format, with separate columns for each year\n",
        "qpd_aggregated <- qpd_summarized %>%\n",
        "  tidyr::pivot_wider(names_from = year, values_from = total_value, names_prefix = \"year_\")\n",
        "\n",
        "# show the final result: qpd_aggregated\n",
        "qpd_aggregated"
      ],
      "id": "0ab907e1-6a1a-46cf-867a-8364cd12d67e"
    }
  ],
  "nbformat": 4,
  "nbformat_minor": 5,
  "metadata": {
    "kernelspec": {
      "name": "ir",
      "display_name": "R",
      "language": "r"
    }
  }
}