Except for JPMorgan Chase, Citigroup, Capital One and Bank of America, the share of minority employees in the total U.S.-based workforce of the leading U.S. banks was less than ** percent. Among the observed banks, JPMorgan Chase had the most diverse workforce, with ** percent of the employees who self-identified were racial minorities. JPMorgan Chase was followed by Citigroup, where the share of people of color was approximately ** percent. Capital One ranked third in terms of racial diversity. Here, the share of non-white employees in the U.S.-based workforce was **** percent. The share was the lowest at PNC Financial Services, where approximately ** percent of the workforce were non-white.

The percentage of local elected officials who are White is higher than the White share of the U.S. population, and this has been the case throughout our country’s history. Yet, few studies have examined the relationship between compensation and candidates’ racial diversity, specifically in small to midsize localities where local elected positions are often volunteer positions. Our research explored whether mayoral and city council candidates are limited to those with enough funds to volunteer, who tend to be Caucasian identifying? Our a priori hypothesis was that paying mayors and city council members is one salient policy solution to encouraging racial diversity among candidates. We analyzed county-archived election data across a sample of 18 small to midsize cities in Oregon that pay their elected officials, noting changes in racial demographics of candidates in the decade prior to implementation of payment and the decade after. We found that, on average, cities saw substantial increases in candidates’ diversity after payment implementation. This article provides straightforward, practical strategies for cities to elect diverse leaders while expanding both the passive and active representation of elected officials.

IntroductionExperts continue to debate how to increase COVID-19 vaccination rates. Some experts advocate for financial incentives. Others argue that financial incentives, especially large ones, will have counterproductive psychological effects, reducing the percent of people who want to vaccinate. Among a racially and ethnically diverse U.S. sample of lower income adults, for whom vaccine uptake has lagged compared with higher income adults, we empirically examine such claims about relatively large and small guaranteed cash payments.MethodsIn 2021, we conducted a randomized, controlled experiment among U.S. residents with incomes below $80,000 who reported being unvaccinated against COVID-19. Study participants were randomized to one of four study arms. In two arms, respondents first learned about a policy proposal to pay $1,000 or $200 to those who received COVID-19 vaccination and were then asked if, given that policy, they would want to vaccinate. In the two other arms, respondents received either an educational message about this vaccine or received no vaccine information and were then asked if they wanted to vaccinate for COVID-19. The primary analyses estimated and compared the overall percentage in each study arm that reported wanting to vaccinate for COVID-19. In other analyses, we estimated and compared these percentages for subgroups of interest, including gender, race/ethnicity, and education.Main resultsAmong 2,290 unvaccinated adults, 79.7% (95%CI, 76.4–83.0%) of those who learned about the proposed $1,000 payment wanted to get vaccinated, compared with 58.9% (95%CI, 54.8–63.0%) in the control condition without vaccine information, a difference of 20 percentage points. Among those who learned of the proposed $200 payment, 74.8% (95% CI, 71.3–78.4%) wanted to vaccinate. Among those who learned only about the safety and efficacy of COVID-19 vaccines, 68.9% (95% CI, 65.1–72.7%) wanted to vaccinate. Findings were consistent across various subgroups.DiscussionDespite several study limitations, the results do not support concerns that the financial incentive policies aimed to increase COVID-19 vaccination would have counterproductive effects. Instead, those who learned about a policy with a large or small financial incentive were more likely than those in the control condition to report that they would want to vaccinate. The positive effects extended to subgroups that have been less likely to vaccinate, including younger adults, those with less education, and racial and ethnic minorities. Financial incentives of $1,000 performed similarly to those offering only $200.

This layer contains information on technology access by Household. Data is from US Census American Community Survey (ACS) 5-year estimates.This layer represents the underlying data for several data visualizations on the Tempe Equity Map.Data visualized as a percent of total population in households in given census tract.Layer includes:Key demographicsTotal Population in Households% Broadband Internet Subscription: American Indian and Alaska Native alone% Broadband Internet Subscription: Asian Alone% Broadband Internet Subscription: Black or African American alone% Broadband Internet Subscription: Native Hawaiian and Other Pacific Islander alone% Broadband Internet Subscription: White Alone% Broadband Internet Subscription: Hispanic or Latino origin% Without an internet Subscription: American Indian and Alaska Native alone% Without an internet Subscription: Asian alone% Without an internet Subscription: Native Hawaiian and Other Pacific Islander alone% Without an internet Subscription: Black or African American Alone% Without an internet Subscription: White Alone% Without an internet Subscription: Hispanic or Latino origin% No computer in household: American Indian and Alaska native alone% No computer in household: Asian alone% No computer in household: Black or African American alone% No computer in household: Native Hawaiian or Pacific Islander% No computer in household: White Alone% No computer in household: Hispanic or Latino originCurrent Vintage: 2018-2022ACS Table(s): S2802 (Not all lines of this ACS table are available in this feature layer.)Census API: Census Bureau's API for American Community SurveyDate of Census update: Dec 15, 2023Data Preparation: Data table downloaded and joined with Census Tract boundaries that are within or adjacent to the City of Tempe boundaryNational Figures: data.census.gov

Personal Financial Management 2.0 Modules Market Outlook

According to our latest research, the global Personal Financial Management 2.0 Modules market size reached USD 3.42 billion in 2024, reflecting a robust adoption trajectory across various sectors. The market is projected to expand at a CAGR of 13.7% from 2025 to 2033, with the forecasted market size reaching USD 10.35 billion by 2033. This impressive growth is primarily driven by the increasing digitalization of financial services, the proliferation of mobile banking, and the growing need for advanced personal finance tools to manage complex financial portfolios.

One of the key growth factors propelling the Personal Financial Management 2.0 Modules market is the rapid evolution of digital banking and fintech ecosystems. As financial institutions and fintech startups race to provide seamless and user-friendly experiences, the demand for innovative personal finance modules has surged. Consumers are now seeking solutions that not only track expenses and budgets but also provide intelligent insights, predictive analytics, and proactive financial recommendations. The integration of artificial intelligence, machine learning, and big data analytics into these modules is enabling more personalized and actionable financial guidance, driving higher user engagement and satisfaction. Furthermore, the rise of open banking APIs and regulatory support for data portability is accelerating the adoption of these advanced financial management tools across both developed and emerging economies.

Another significant driver is the shift in consumer behavior towards proactive financial management. The aftermath of global economic disruptions, such as the COVID-19 pandemic, has heightened awareness of personal financial health. Individuals and businesses alike are increasingly leveraging Personal Financial Management 2.0 Modules to gain real-time visibility into their finances, optimize spending, and make informed investment decisions. This trend is particularly pronounced among younger, tech-savvy demographics who demand intuitive mobile solutions and seamless integration with other financial services. The growing emphasis on financial literacy and wellness programs by employers and educational institutions is also augmenting market growth, as more users recognize the value of comprehensive financial planning and management tools.

Additionally, the ongoing digital transformation in the financial services industry is fostering collaboration between traditional banks and fintech innovators. Financial institutions are investing heavily in the development and deployment of next-generation personal finance modules to retain and attract customers. These platforms are being designed to offer holistic financial management, combining budgeting, investment, debt management, and personalized financial planning within a single interface. The emergence of embedded finance, where non-financial companies integrate financial services into their offerings, is further expanding the addressable market. As a result, the competitive landscape is becoming increasingly dynamic, with established players and new entrants vying to differentiate their offerings through superior technology, user experience, and value-added services.

From a regional perspective, North America continues to lead the global Personal Financial Management 2.0 Modules market, accounting for over 38% of the total market share in 2024. The region's dominance is attributed to a mature digital banking infrastructure, high smartphone penetration, and a strong culture of financial innovation. However, Asia Pacific is emerging as the fastest-growing region, with a CAGR exceeding 16.2% during the forecast period. The rapid adoption of digital payment solutions, rising middle-class population, and supportive government policies are fueling market expansion in countries such as China, India, and Southeast Asia. Meanwhile, Europe and Latin America are also witnessing steady growth, driven by regulatory initiatives and increasing consumer awareness of financial wellness.

This dataset contains information related to individuals' financial and demographic characteristics, which can be useful for various analyses, such as credit scoring, customer segmentation, or financial behavior studies. The dataset includes the following columns:

• ID: A unique identifier for each individual.
• Income: The individual's annual income (in thousands of dollars).
• Limit: The credit limit assigned to the individual's credit card account.
• Rating: A credit rating score for the individual.
• Cards: The number of credit cards the individual possesses.
• Age: The age of the individual.
• Education: The number of years of education completed by the individual.
• Gender: The gender of the individual (Male/Female).
• Student: Whether the individual is a student (Yes/No).
• Married: The marital status of the individual (Yes/No).
• Ethnicity: The ethnic background of the individual.
• Balance: The current balance on the individual's credit card account. ### Key Insights
• Income and Credit Limit: Examining the relationship between income levels and the assigned credit limits.
• Credit Rating: Understanding how various factors such as income, number of cards, and education influence credit ratings.
• Demographic Analysis: Analyzing differences in financial behavior and credit characteristics across different genders, marital statuses, and ethnic backgrounds.
• Age and Financial Behavior: Investigating how financial behavior, such as balance and credit card usage, changes with age. ### Potential Applications
• Credit Risk Modeling: Develop models to predict credit risk based on demographic and financial data.
• Customer Segmentation: Identify different customer segments to tailor financial products and marketing strategies.
• Financial Education Programs: Design targeted financial education programs for specific demographics to improve financial literacy and behavior.

This dataset provides a comprehensive view of various factors that can influence an individual's financial behavior and creditworthiness, making it a valuable resource for financial analysis and modeling.

The Survey of Consumer Finances (SCF) is normally a triennial cross-sectional survey of U.S. families. The survey data include information on families' balance sheets, pensions, income, and demographic characteristics.

This analysis presents a rigorous exploration of financial data, incorporating a diverse range of statistical features. By providing a robust foundation, it facilitates advanced research and innovative modeling techniques within the field of finance.

Allogene's (ALLO) Race to Cure Cancer: A Stock Forecast

Financial data:

• Historical daily stock prices (open, high, low, close, volume)

• Fundamental data (e.g., market capitalization, price to earnings P/E ratio, dividend yield, earnings per share EPS, price to earnings growth, debt-to-equity ratio, price-to-book ratio, current ratio, free cash flow, projected earnings growth, return on equity, dividend payout ratio, price to sales ratio, credit rating)

• Technical indicators (e.g., moving averages, RSI, MACD, average directional index, aroon oscillator, stochastic oscillator, on-balance volume, accumulation/distribution A/D line, parabolic SAR indicator, bollinger bands indicators, fibonacci, williams percent range, commodity channel index)

Machine learning features:

• Feature engineering based on financial data and technical indicators

• Sentiment analysis data from social media and news articles

• Macroeconomic data (e.g., GDP, unemployment rate, interest rates, consumer spending, building permits, consumer confidence, inflation, producer price index, money supply, home sales, retail sales, bond yields)

Potential Applications:

• Stock price prediction

• Portfolio optimization

• Algorithmic trading

• Market sentiment analysis

• Risk management

Use Cases:

• Researchers investigating the effectiveness of machine learning in stock market prediction

• Analysts developing quantitative trading Buy/Sell strategies

• Individuals interested in building their own stock market prediction models

• Students learning about machine learning and financial applications

Additional Notes:

• The dataset may include different levels of granularity (e.g., daily, hourly)

• Data cleaning and preprocessing are essential before model training

• Regular updates are recommended to maintain the accuracy and relevance of the data

IntroductionA large body of research has established a consensus that racial discrimination in CV screening occurs and persists. Nevertheless, we still know very little about how recruiters look at the CV and how this is connected to the discriminatory patterns. This article examines the way recruiters view and select CVs and how they reason about their CV selection choices, as a first step in unpacking the patterns of hiring discrimination. Specifically, we explore how race and ethnicity signaled through the CV matter, and how recruiters reason about the choices they make.MethodsWe recorded data from 40 respondents (20 pairs) who are real-life recruiters with experiences in recruitment of diverse employees in three large Swedish-based firms in the finance and retail sector in two large cities. The participating firms all value diversity, equity and inclusion in their recruitment. Their task was to individually rate 10 fictious CVs where race (signaled by face image) and ethnicity (signaled by name) were systematically manipulated, select the top three candidates, and then discuss their choices in pairs to decide on a single top candidate. We examined whether respondents’ choices were associated with the parts of the CV they looked at, and how they reasoned and justified their choices through dialog.ResultsOur results show that non-White CVs were rated higher than White CVs. While we do not observe any statistically significant differences in the ratings between different racial groups, we see a statistically significant preference for Chinese over Iraqi names. There were no significant differences in time spent looking at the CV across different racial groups, but respondents looked longer at Polish names compared to Swedish names when presented next to a White face. The dialog data reveal how respondents assess different CVs by making assumptions about the candidates’ job and organizational fit through limited information on the CVs, especially when the qualifications of the candidates are evaluated to be equal.

The Distributional Financial Accounts (DFAs) provide a quarterly measure of the distribution of U.S. household wealth since 1989, based on a comprehensive integration of disaggregated household-level wealth data with official aggregate wealth measures. The data set contains the level and share of each balance sheet item on the Financial Accounts' household wealth table (Table B.101.h), for various sub-populations in the United States. In our core data set, aggregate household wealth is allocated to each of four percentile groups of wealth: the top 1 percent, the next 9 percent (i.e., 90th to 99th percentile), the next 40 percent (50th to 90th percentile), and the bottom half (below the 50th percentile). Additionally, the data set contains the level and share of aggregate household wealth by income, age, generation, education, and race. The quarterly frequency makes the data useful for studying the business cycle dynamics of wealth concentration--which are typically difficult to observe in lower-frequency data because peaks and troughs often fall between times of measurement. These data will be updated about 10 or 11 weeks after the end of each quarter, making them a timely measure of the distribution of wealth.

Author's Note 2019/04/20: Revisiting this project, I recently discovered the incredibly comprehensive API produced by the Urban Institute. It achieves all of the goals laid out for this dataset in wonderful detail. I recommend that users interested pay a visit to their site.

Context

This dataset is designed to bring together multiple facets of U.S. education data into one convenient CSV (states_all.csv).

Contents

• states_all.csv: The primary data file. Contains aggregates from all state-level sources in one CSV.

• output_files/states_all_extended.csv: The contents of states_all.csv with additional data related to race and gender.

Column Breakdown

Identification

• PRIMARY_KEY: A combination of the year and state name.
• YEAR
• STATE

Enrollment

A breakdown of students enrolled in schools by school year.

• GRADES_PK: Number of students in Pre-Kindergarten education.

• GRADES_4: Number of students in fourth grade.

• GRADES_8: Number of students in eighth grade.

• GRADES_12: Number of students in twelfth grade.

• GRADES_1_8: Number of students in the first through eighth grades.

• GRADES 9_12: Number of students in the ninth through twelfth grades.

• GRADES_ALL: The count of all students in the state. Comparable to ENROLL in the financial data (which is the U.S. Census Bureau's estimate for students in the state).

The extended version of states_all contains additional columns that breakdown enrollment by race and gender. For example:

• G06_A_A: Total number of sixth grade students.

• G06_AS_M: Number of sixth grade male students whose ethnicity was classified as "Asian".

• G08_AS_A_READING: Average reading score of eighth grade students whose ethnicity was classified as "Asian".

The represented races include AM (American Indian or Alaska Native), AS (Asian), HI (Hispanic/Latino), BL (Black or African American), WH (White), HP (Hawaiian Native/Pacific Islander), and TR (Two or More Races). The represented genders include M (Male) and F (Female).

Financials

A breakdown of states by revenue and expenditure.

• ENROLL: The U.S. Census Bureau's count for students in the state. Should be comparable to GRADES_ALL (which is the NCES's estimate for students in the state).

• TOTAL REVENUE: The total amount of revenue for the state.

  • FEDERAL_REVENUE
  • STATE_REVENUE
  • LOCAL_REVENUE

• TOTAL_EXPENDITURE: The total expenditure for the state.

  • INSTRUCTION_EXPENDITURE

  • SUPPORT_SERVICES_EXPENDITURE

  • CAPITAL_OUTLAY_EXPENDITURE

  • OTHER_EXPENDITURE

Academic Achievement

A breakdown of student performance as assessed by the corresponding exams (math and reading, grades 4 and 8).

• AVG_MATH_4_SCORE: The state's average score for fourth graders taking the NAEP math exam.

• AVG_MATH_8_SCORE: The state's average score for eight graders taking the NAEP math exam.

• AVG_READING_4_SCORE: The state's average score for fourth graders taking the NAEP reading exam.

• AVG_READING_8_SCORE: The state's average score for eighth graders taking the NAEP reading exam.

Data Processing

The original sources can be found here:

# Enrollment
https://nces.ed.gov/ccd/stnfis.asp
# Financials
https://www.census.gov/programs-surveys/school-finances/data/tables.html
# Academic Achievement
https://www.nationsreportcard.gov/ndecore/xplore/NDE

Data was aggregated using a Python program I wrote. The code (as well as additional project information) can be found [here][1].

Methodology Notes

• Spreadsheets for NCES enrollment data for 2014, 2011, 2010, and 2009 were modified to place key data on the same sheet, making scripting easier.

• The column 'ENROLL' represents the U.S. Census Bureau data value (financial data), while the column 'GRADES_ALL' represents the NCES data value (demographic data). Though the two organizations correspond on this matter, these values (which are ostensibly the same) do vary. Their documentation chalks this up to differences in membership (i.e. what is and is not a fourth grade student).

• Enrollment data from NCES has seen a number of changes across survey years. One of the more notable is that data on student gender does not appear to have been collected until 2009. The information in states_all_extended.csv reflects this.

• NAEP test score data is only available for certain years

• The current version of this data is concerned with state-level patterns. It is the author's hope that future versions will allow for school district-level granularity.

Acknowledgements

Data is sourced from the U.S. Census Bureau and the National Center for Education Statistics (NCES).

Licensing Notes

The licensing of these datasets state that it must not be us...

The Survey of Consumer Finances (SCF) dataset, provided by the Federal Reserve, offers comprehensive insights into the financial condition of U.S. households. This dataset is invaluable for researchers, policymakers, and analysts interested in understanding consumer behavior, wealth distribution, and economic trends in the United States.

The SCF dataset includes detailed information on household income, assets, liabilities, and various demographic characteristics. It is collected every three years and serves as a crucial resource for analyzing the financial well-being of American families.

Key Features: Income Data: Information on various sources of income, including wages, investments, and government assistance. Asset Ownership: Detailed accounts of household assets, such as real estate, retirement accounts, stocks, and other investments. Liabilities:Comprehensive details on household debts, including mortgages, credit card debts, and student loans. Demographics: Data covering age, education, race, and family structure, allowing for nuanced analysis of financial trends across different segments of the population.

Use Cases: Economic research and analysis, Policy formulation and assessment, Understanding wealth inequality, Consumer behavior studies

Citing the Dataset:

When using this dataset in your research, please ensure to cite the Federal Reserve Board and the SCF as the original source.

Note: The dataset is intended for educational and research purposes. Users are encouraged to adhere to ethical guidelines when analyzing and interpreting the data.

Abstract Purpose: This study aims to examine the influence of board diversity on the quality of CSR disclosure (QCSR) and propose that this relationship is patterned differently in different contexts and nations, due to their distinctive characteristics. Design/methodology/approach: The resource-based view (RBV) theory is used to evaluate the hypothesized relationship through an empirical investigation of 64 Pakistani financial firms, by applying a random-effects regression and the generalized method of moments (GMM).] Findings: The findings indicate that age, gender, educational level, and educational background diversities positively influence QCSR disclosure. However, nationality, ethnic, and tenure diversities have no significant relationship with QCSR disclosure. The results were further checked by a robust regression and sensitivity analysis. Originality/value: Using RBV theory, this paper provides an additional contribution concerning the role played by board diversity in a firm’s strategic performance, particularly CSR disclosure. The article contributes to the literature by finding that there is no unanimous rule for board diversity supporting CSR, due to the unique characteristics of different jurisdictions and institutional contexts.

SABAL - Small Area Business and Labour Data is a compendium of independent data sources brought together in one database by Statistics Canada. As a compendium, it is not a fully integrated system, therefore, dates and geographic areas covered vary between data sets selected. SABAL combines a wide variety of economic and social statistics, and provides coverage of approximately 140 urban areas and 72 economic regions, in addition to Canada, the Provinces and Territories. Some data are not available at all geographic levels. SABAL also includes metadata on each of these data sources. The business sources included are: Business Small Area File (based on Revenue Canada administrative data), Retail Trade, Building Permits, Housing Starts from CMHC, Survey of Manufacturing, Motor Vehicle Registrations, Business Counts, Consumer Price Index, and Tourism. The social sources included are: Census of Population, Small Area Administrative Data (Taxfiler Data), Labour Force Survey, Household Facilities and Equipment Survey, Education, Training, Justice, Population Projections, Family Expenditures, and Consumer Finances Survey.

In this file there are statistics for a number of variables broken down by Malmö’s different areas over time.

Source

Unless otherwise stated, the statistics in this database are retrieved from Statistics Sweden’s (SCB) regional database, Skånedatabasen or from Statistics Sweden’s area statistics database (OSDB). The Skåne database and OSDB show data from several different sources that Statistics Sweden has compiled on a geographical level. The statistics only cover persons who are part of the population registered in the population. Therefore, persons without a residence permit, such as asylum seekers, and persons who simply have not registered in the municipality are not included. Statistics Sweden does not provide statistics on which language residents speak, which religion you belong to or what ethnicity or political views you have. Therefore, such data is not available here either. However, the Electoral Authority reports election results per constituency on its website val.se. There are statistics from the last election as well as several previous elections available. Please note, however, that the constituencies do not necessarily follow the division of the city made here.

Update

The data is updated every spring as Statistics Sweden releases the figures to the municipality. Most variables are available for the year before. However, income and employment data are released with another year’s backlog. Unless otherwise stated, the date of measurement is 31 December of each year.

Geographical breakdown

Unless otherwise stated, the data is available for Malmö as a whole and broken down into urban areas (5 pieces), districts (10 pieces) and subareas (136 pieces). In addition to these, there is a residual post that contains the people who are not written in a specific place in the municipality, have protected identity and more. These people are also part of the total. In several of the subareas there are no or only a few registered population registers. Therefore, no data are reported for these areas. Examples of such sub-areas are parks such as Pildammsparken and Kroksbäcksparken and industrial areas such as Fosieby Industriområde and Spillepengen.

Privacy clearance

In order to protect the identity of individuals, the data is confidentially audited. This means that small values are suppressed, i.e. replaced by empty cells. However, the values are included in summaries. In general, the following rules apply:

• No statistics are reported for geographical areas with very few housing.
• No cells with fewer than 5 individuals are reported. For data classified as sensitive (e.g. income and country of birth), larger values can also be suppressed.
• In cases where a subcategory (e.g. a training category) is too small to be accounted for, all categories are often suppressed.

Please use the numbers, but use “City Office, Malmö City” as the source.

Analysis of mean financial stress scores across religious affiliations in a predominantly Black sample.

Project Rosalind CBDC APIs Market Outlook

According to our latest research, the Global Project Rosalind CBDC APIs market size was valued at $1.2 billion in 2024 and is projected to reach $9.7 billion by 2033, expanding at a robust CAGR of 26.4% during 2024–2033. The primary factor fueling this rapid growth is the accelerated adoption of Central Bank Digital Currencies (CBDCs) by governments and financial institutions worldwide, which has created a pressing need for secure, scalable, and interoperable API solutions. As central banks and commercial entities race to modernize their payment infrastructures, Project Rosalind CBDC APIs are emerging as the backbone for next-generation digital currency ecosystems, enabling seamless integration, enhanced security, and innovative financial products across diverse platforms and stakeholders.

Regional Outlook

North America holds the largest share of the Project Rosalind CBDC APIs market, accounting for approximately 38% of the global revenue in 2024. This dominance is attributed to the region’s mature financial technology landscape, early adoption of digital currency initiatives, and progressive regulatory frameworks that encourage innovation in digital payments. The presence of major API platform providers, coupled with strong collaboration between central banks and private sector players, has fostered a conducive environment for the development and deployment of CBDC APIs. Additionally, North America benefits from well-established cybersecurity infrastructure and significant investments in cloud-based solutions, further reinforcing its leadership position in the global market.

The Asia Pacific region is projected to be the fastest-growing market for Project Rosalind CBDC APIs, with a forecasted CAGR exceeding 30% between 2025 and 2033. This remarkable growth is driven by substantial government investments in digital currency pilots, particularly in countries such as China, India, and Singapore, where central banks are rapidly advancing CBDC projects. The region’s large unbanked population, increasing smartphone penetration, and robust fintech ecosystem are catalyzing the demand for innovative payment solutions powered by CBDC APIs. Furthermore, Asia Pacific’s proactive approach to regulatory sandboxes and digital identity initiatives is accelerating the integration of these APIs into mainstream financial services, positioning the region as a global leader in digital currency adoption.

Emerging economies in Latin America and Middle East & Africa are witnessing gradual adoption of Project Rosalind CBDC APIs, albeit with unique challenges. In these regions, the focus is on leveraging CBDC APIs to enhance financial inclusion, streamline cross-border remittances, and reduce transaction costs. However, issues such as limited digital infrastructure, regulatory uncertainty, and varying levels of technological literacy pose significant hurdles to widespread adoption. Localized demand is being shaped by government-led pilot programs and partnerships with international organizations, but the pace of implementation remains contingent on policy clarity and investment in digital infrastructure. Despite these challenges, the long-term outlook remains positive, with increasing interest from central banks and fintech innovators seeking to bridge the digital divide.

Report Scope

This dataset was introduced in a competition on Zindi to challenge data professionals to predict whether members of the test population would be earning below or above $50,000 based on the variables taken into account in the analysis.

Analysis Objective.

The objective of this challenge is to create a machine learning model to predict whether an individual earns above or below a certain amount.

This solution can potentially reduce the cost and improve the accuracy of monitoring key population indicators such as income level in between census years. This information will help policymakers to better manage and avoid income inequality globally.

Data background.

This data has been collected from a random population.

There are ~200 000 individuals in train and ~100 000 individuals in the test file.

The train & test data will be used to create a machine learning model to predict if an individual earns above 50 000 of a specific currency.

The key variables are as follows: * Age. * Gender. * Education. * Class. * Education institute. * Marital status. * Race. * Is hispanic. * Employment commitment. * Unemployment reason. * Employment state. * Wage per hour. * Is part of labor union. * Working week per year. * Industry code. * Main Industry code. * Occupation code. * Main Occupation code. * Total employed. * Household stat. * Household summary. * Under 18 family. * Veterans adminquestionnaire. * Veteran benefit. * Tax status. * Gains. * Losses. * Stocks status. * Citizenship. * Migration year. * Country of birth own. * Country of birth father. * Country of birth mother. * Migration code change in msa. * Migration previous sunbelt. * Migration code move within registration. * Migration code change in registration. * Residence 1 year ago. * Old residence registration. * Old residence state. * Importance of record. * Income above limit.

Based on the variables set up and the data target requirements, the analysis can be assumed to be based on 20th century American population data where the median income was about $ 50,000.

Why Predict Income?

Income prediction extracts insights from individual and population-level data as it offers the ability to forecast income levels, assess financial risks, target marketing campaigns, and inform crucial decision-making in diverse spheres. However, ethical considerations, potential biases, and data privacy concerns demand careful attention alongside its undeniable benefits.

Applications in Specific Industries:

Finance:

• Credit scoring and loan approval: Predicting income to assess risk and determine loan eligibility.
• Fraud detection: Identifying suspicious financial activity based on income patterns.
• Targeted marketing: Tailoring financial products and services to individuals based on predicted income.
• Wealth management: Personalized investment strategies based on income potential and risk tolerance.

Healthcare:

• Predicting healthcare costs and resource allocation based on patient income and demographics.
• Identifying individuals at risk of financial hardship due to medical bills.
• Developing targeted health insurance plans based on income segments.
• Assessing eligibility for government healthcare programs.

Marketing and Retail:

• Customer segmentation and targeted advertising based on predicted income groups.
• Product pricing strategies tailored to different income segments.
• Predicting customer lifetime value and optimizing marketing spend.
• Identifying potential high-value customers based on income potential.

Human Resources:

• Salary benchmarking and compensation planning based on industry standards and predicted income trends.
• Predicting employee turnover risk based on job satisfaction and income-related factors.
• Identifying and attracting top talent by offering competitive compensation packages.
• Optimizing talent management strategies based on income potential and skillsets.

Public Policy:

• Predicting tax revenue and allocating government resources based on income distribution.
• Identifying individuals in need of social assistance programs based on income poverty.
• Evaluating the effectiveness of government policies aimed at income inequality.
• Designing progressive taxation systems based on predicted income levels.

Challenges and Considerations:

• Data privacy and security concerns related to income information.
• Potential for biases and discrimination based on income prediction models.
• Explainability and transparency of income prediction algorithms.
• The impact of income prediction on individual behavior and choices.

Resource Management (RM) was the system used throughout DWP from 2006 to carry out all its transactional human resources, payroll, finance and procurement functions. RM held data on over 100,000 staff on behalf of all clients (DWP, Cabinet Office, Department for Education, Health and Safety Executive). This data comprised the full set of protected personal information: name; address; National Insurance number, bank account details; date of birth. The data held also encompassed data that could be considered sensitive such as disability, sexual orientation, ethnicity; previous names, tax details, next of kin, sick absence dates and reasons, union membership. Note some elements of this data set were supplied on a voluntary basis and cannot be guaranteed to be complete for all employees. The system also contained, payroll records, pay and grade - current and history, allowances, expenses and overtime claims, advances and loans; payments and recoveries, diversity, learning & development activity, educational/professional qualifications, performance appraisal records and the purchase of good and services. As part of the procurement process RM held the details of requisitions raised (approximately 14,000 a year for all clients), purchase orders raised (some 11,500 per year for all clients), and associated supplier details and order values, invoices received and payments made, including supplier bank details and values (approximately 4,200 Bankers' Automated Clearing Services payments per year). The Purchase to Pay service line also dealt with Accounts Receivable transactions. This dealt with approx. 96,000 receipts a year with the system holding details of individual debtors, the balanced owed, nature of debt and payments received. RM held the accounting structures for each client and records of all financial transactions, including manual and automatically created journals with values, descriptions, dates and audit reference data. Some 32,000 journals were processed per year. RM also contained records for Fixed Assets held for all clients. This covered classification information, description of the asset, initial financial value, depreciation and current value of the asset. Transaction details were present for additions, deletions, revaluations, annual depreciation (approximately 4730 Fixed Asset related transactions per year). VAT transactions also formed part of the accounting data with VAT payment and reconciliation data for clients being held. The Resource Management System was in place from 2006 for DWP, April 2009 for Cabinet Office, and September 2009 for Department for Education. The data covered all parts of the UK and as well as the 11,300 current records there was historical data.