raw_code_python

import requests
import xml.etree.ElementTree as ET
import matplotlib.pyplot as plt
import pandas as pd
import ipywidgets as widgets
from IPython.display import display, clear_output
from scipy.stats import ttest_ind
import csv

income_level_codes = {
    'Low income': "LIC",
    'Lower middle income': "LMC",
    'Upper middle income': "UMC"
}

# fetch income groups
# per page = 300 to get all countries in one request
url = "https://api.worldbank.org/v2/country?format=xml&per_page=300"
response = requests.get(url)

root = ET.fromstring(response.content)

# Define the namespace
namespace = {'wb': 'http://www.worldbank.org'}

# Extract country codes and income levels
country_dict = {}
country_dict['LIC'] = []
country_dict['LMC'] = []
country_dict['UMC'] = []

for country in root.findall('wb:country', namespace):
    country_id = country.get('id')
    income_level = country.find('wb:incomeLevel', namespace).text

    for description, code in income_level_codes.items():
        if description == income_level:
            country_dict[code].append(country_id)

# fetch indicators
# List of indicators (for diarrhea prevalence and treatment in poorest vs richest quintiles)
indicators = {
    "Prevalence": {
        "Diarrhea prevalence of children under the age of 5 - Poorest quintile": "SH.STA.DIRH.Q1.ZS",
        "Diarrhea prevalence of children under the age of 5 - Richest quintile": "SH.STA.DIRH.Q5.ZS",
    },
    "Treatment": {
        "Diarrhea treatment of children under the age of 5 - Poorest quintile": "SH.STA.ORHF.Q1.ZS",
        "Diarrhea treatment of children under the age of 5 - Richest quintile": "SH.STA.ORHF.Q5.ZS",
    },
}


# fetch the most recent year of data for a specific country and indicator
def fetch_latest_data(indicator):
    # mrnev=1 get most recent non-empty
    url = f"https://api.worldbank.org/v2/country/all/indicator/{indicator}?mrnev=1&format=xml&per_page=300"
    response = requests.get(url)
    if response.status_code == 200:
        try:
            data = response.content
            return ET.fromstring(response.content)
        except Exception as e:
            print(f"Error processing data for {indicator}: {e}")
    return None


# Fetch data for each indicator and country
indicator_dict = {}
for income_group, income_code in income_level_codes.items():
    indicator_dict[income_code] = {}

    for prev_treat, indicator_codes in indicators.items():
        indicator_dict[income_code][prev_treat] = {}

        for indicator_name, indicator_code in indicator_codes.items():
            indicator_dict[income_code][prev_treat][indicator_code] = {}

            record = fetch_latest_data(indicator_code)
            for data in record.findall("wb:data", namespace):
                iso3code = data.find("wb:countryiso3code", namespace).text
                value = data.find("wb:value", namespace).text

                for country_income_group, country_list in country_dict.items():
                    if income_code == country_income_group:
                        for country in country_list:
                            if country == iso3code:
                                indicator_dict[income_code][prev_treat][indicator_code][iso3code] = value

# Function to save dictionary data to a CSV file
def save_data_to_csv(data, filename="data.csv"):
    with open(filename, mode="w", newline="") as file:
        writer = csv.writer(file)
        # Write the header
        writer.writerow(["Indicator", "Country", "Percentage"])

        # Write data rows
        for indicator, countries in data.items():
            for country, percentage in countries.items():
                writer.writerow([indicator, country, percentage])
    print(f"Data saved to {filename}")

def get_indicator_data(selected_income_group, selected_indicator):
    rich_vs_poor = {}

    for _, indicator_code in indicators[selected_indicator].items():
        rich_vs_poor[indicator_code] = indicator_dict[selected_income_group][selected_indicator][indicator_code]

    return rich_vs_poor

def calculate_t_test(data):
    # Extract values as floats
    vals = []
    for subdict in data.values():
        vals.append(list(map(float, subdict.values())))

    if len(vals) != 2:
        raise ValueError("The data should contain exactly two groups for a t-test.")

    # Perform a two-sample independent t-test
    return ttest_ind(vals[0], vals[1])

# Create GUI with ipywidgets
income_group_selector = widgets.Dropdown(
    options=list(income_level_codes.keys()),
    description='Income Group:',
    style={'description_width': 'initial'}
)

indicator_selector = widgets.Dropdown(
    options=list(indicators.keys()),
    description='Indicator:',
    style={'description_width': 'initial'}
)

fetch_button = widgets.Button(
    description="Get Data",
    button_style='success',
    tooltip='Display visualization',
    icon='search'
)

output_area = widgets.Output()

# Fetch and visualize data on button click
def on_fetch_button_click(b):
    with output_area:
        clear_output()
        selected_income_group = income_group_selector.value
        selected_income_code = income_level_codes[selected_income_group]
        selected_indicator_name = indicator_selector.value
        # selected_indicator_code = indicators[selected_indicator_name]

        # Display boxplot
        data = get_indicator_data(selected_income_code, selected_indicator_name)

        # Prepare boxplot data
        values = [list(map(float, subdict.values())) for subdict in data.values()]
        labels = []
        for key in data.keys():
            if 'Q1' in key:
                labels.append('poorest quintile')
            if 'Q5' in key:
                labels.append('richest quintile')

        # Create the boxplot
        fig, ax = plt.subplots()
        ax.boxplot(values, labels=labels)
        ax.set_ylabel("Percentage (%)")
        ax.set_title(f"{selected_indicator_name} of diarrhea in {selected_income_group.lower()} countries")
        plt.show()

        # Save CSV data
        download_button = widgets.Button(
            description="Download CSV",
            button_style='info'
        )

        def on_download_button_click(d):
            save_data_to_csv(data, filename = f"{selected_income_code}_{selected_indicator_name}.csv")

        download_button.on_click(on_download_button_click)
        display(download_button)

        # ttest
        t_test_button = widgets.Button(
            description='t test',
            button_style='info',
            tooltip='Calculate t statistic'
        )

        def on_ttest_button_click(d):
            t_stat, p_value = calculate_t_test(data)

            # Print the results
            with output_area:
                display(f"T-statistic: {t_stat}")
                display(f"P-value: {p_value}")

                # Interpretation
                alpha = 0.05
                if p_value < alpha:
                    display("The difference between the two groups is statistically significant.")
                else:
                    display("The difference between the two groups is not statistically significant.")

        t_test_button.on_click(on_ttest_button_click)
        display(t_test_button)


fetch_button.on_click(on_fetch_button_click)

# Display widgets
display(widgets.VBox([income_group_selector, indicator_selector, fetch_button, output_area]))