Software Job Market Comparison Graph

sjmcg.py

import pandas as pd
import matplotlib.pyplot as plt

# load
# https://fred.stlouisfed.org/series/IHLIDXUSTPHUMARESO
df_hr = pd.read_csv('IHLIDXUSTPHUMARESO.csv', parse_dates=['DATE'])
# https://fred.stlouisfed.org/series/IHLIDXUSTPSOFTDEVE
df_software = pd.read_csv('IHLIDXUSTPSOFTDEVE.csv', parse_dates=['DATE'])
# https://fred.stlouisfed.org/series/IHLIDXUSTPPROJMANA
df_pm = pd.read_csv('IHLIDXUSTPPROJMANA.csv', parse_dates=['DATE'])
# https://fred.stlouisfed.org/series/IHLIDXUSTPSALE
df_sales = pd.read_csv('IHLIDXUSTPSALE.csv', parse_dates=['DATE'])
# https://fred.stlouisfed.org/series/IHLIDXUSTPRETA
df_retail = pd.read_csv('IHLIDXUSTPRETA.csv', parse_dates=['DATE'])
# https://fred.stlouisfed.org/series/IHLIDXUSTPITOPHE
df_it_ops = pd.read_csv('IHLIDXUSTPITOPHE.csv', parse_dates=['DATE'])

def adjust_values(df, column_name):
    initial_value = df[column_name].iloc[0]
    df['Adjusted Value'] = df[column_name] - (initial_value - 100)

adjust_values(df_hr, 'IHLIDXUSTPHUMARESO')
adjust_values(df_software, 'IHLIDXUSTPSOFTDEVE')
adjust_values(df_pm, 'IHLIDXUSTPPROJMANA')
adjust_values(df_sales, 'IHLIDXUSTPSALE')
adjust_values(df_retail, 'IHLIDXUSTPRETA')
adjust_values(df_it_ops, 'IHLIDXUSTPITOPHE')

plt.figure(figsize=(14, 8))

plt.plot(df_hr['DATE'], df_hr['Adjusted Value'], color='b', label='HR Job Postings', linewidth=1)
end_hr_value = df_hr['Adjusted Value'].iloc[-1]
lowest_hr_value = df_hr['Adjusted Value'].min()
percent_diff_lowest_hr = ((end_hr_value - lowest_hr_value) / lowest_hr_value) * 100
percent_diff_100_hr = ((end_hr_value - 100) / 100) * 100
plt.text(df_hr['DATE'].iloc[-1], end_hr_value,
         f'{end_hr_value:.2f}\n({percent_diff_lowest_hr:.2f}% from lowest, {percent_diff_100_hr:.2f}% from 100)',
         color='b', fontsize=9, ha='right')

plt.plot(df_software['DATE'], df_software['Adjusted Value'], color='r', label='Software Dev Job Postings', linewidth=1)
end_software_value = df_software['Adjusted Value'].iloc[-1]
lowest_software_value = df_software['Adjusted Value'].min()
percent_diff_lowest_software = ((end_software_value - lowest_software_value) / lowest_software_value) * 100
percent_diff_100_software = ((end_software_value - 100) / 100) * 100
plt.text(df_software['DATE'].iloc[-1], end_software_value,
         f'{end_software_value:.2f}\n({percent_diff_lowest_software:.2f}% from lowest, {percent_diff_100_software:.2f}% from 100)',
         color='r', fontsize=9, ha='right')

plt.plot(df_pm['DATE'], df_pm['Adjusted Value'], color='g', label='Project Management Job Postings', linewidth=1)
end_pm_value = df_pm['Adjusted Value'].iloc[-1]
lowest_pm_value = df_pm['Adjusted Value'].min()
percent_diff_lowest_pm = ((end_pm_value - lowest_pm_value) / lowest_pm_value) * 100
percent_diff_100_pm = ((end_pm_value - 100) / 100) * 100
plt.text(df_pm['DATE'].iloc[-1], end_pm_value,
         f'{end_pm_value:.2f}\n({percent_diff_lowest_pm:.2f}% from lowest, {percent_diff_100_pm:.2f}% from 100)',
         color='g', fontsize=9, ha='right')

plt.plot(df_sales['DATE'], df_sales['Adjusted Value'], color='m', label='Sales Job Postings', linewidth=1)
end_sales_value = df_sales['Adjusted Value'].iloc[-1]
lowest_sales_value = df_sales['Adjusted Value'].min()
percent_diff_lowest_sales = ((end_sales_value - lowest_sales_value) / lowest_sales_value) * 100
percent_diff_100_sales = ((end_sales_value - 100) / 100) * 100
plt.text(df_sales['DATE'].iloc[-1], end_sales_value,
         f'{end_sales_value:.2f}\n({percent_diff_lowest_sales:.2f}% from lowest, {percent_diff_100_sales:.2f}% from 100)',
         color='m', fontsize=9, ha='right')

plt.plot(df_retail['DATE'], df_retail['Adjusted Value'], color='c', label='Retail Job Postings', linewidth=1)
end_retail_value = df_retail['Adjusted Value'].iloc[-1]
lowest_retail_value = df_retail['Adjusted Value'].min()
percent_diff_lowest_retail = ((end_retail_value - lowest_retail_value) / lowest_retail_value) * 100
percent_diff_100_retail = ((end_retail_value - 100) / 100) * 100
plt.text(df_retail['DATE'].iloc[-1], end_retail_value,
         f'{end_retail_value:.2f}\n({percent_diff_lowest_retail:.2f}% from lowest, {percent_diff_100_retail:.2f}% from 100)',
         color='c', fontsize=9, ha='right')

plt.plot(df_it_ops['DATE'], df_it_ops['Adjusted Value'], color='purple', label='IT Operations', linewidth=1)
end_it_ops_value = df_it_ops['Adjusted Value'].iloc[-1]
lowest_it_ops_value = df_it_ops['Adjusted Value'].min()
percent_diff_lowest_it_ops = ((end_it_ops_value - lowest_it_ops_value) / lowest_it_ops_value) * 100
percent_diff_100_it_ops = ((end_it_ops_value - 100) / 100) * 100
plt.text(df_it_ops['DATE'].iloc[-1], end_it_ops_value,
         f'{end_it_ops_value:.2f}\n({percent_diff_lowest_it_ops:.2f}% from lowest, {percent_diff_100_it_ops:.2f}% from 100)',
         color='purple', fontsize=9, ha='right')

# horiz indicator
plt.axhline(y=100, color='k', linestyle='--', linewidth=1, label='Reference Line (100)')

# plot
plt.title('Job Postings and IT Operations Comparison (Adjusted)')
plt.xlabel('Date')
plt.ylabel('Value')
plt.grid(True)
plt.xticks(rotation=45)
plt.legend()
plt.tight_layout()
plt.show()