Predictive model for recruiting process in New York City
No Thumbnail Available
URL
Journal Title
Journal ISSN
Volume Title
School of Business |
Master's thesis
Authors
Date
2023
Department
Major/Subject
Mcode
Degree programme
Information and Service Management (ISM)
Language
en
Pages
64
Series
Abstract
The workforce is an important component of any economy, and knowing future trends and demands can help politicians, educators, and companies prepare for future occupations. This study focuses on predictive models that help workforce forecasting in New York City (NYC) utilizing job open data. The study's goal is to provide insights into present and future employment market trends in NYC, specifically in terms of in-demand skills and jobs. It also intends to make recommendations on how to overcome the skills deficit between the available workforce and the demand for skills, as well as to present competitive salary packages to attract talent. To achieve the research objectives, job opening data from NYC was analyzed and interpreted to identify and forecast future demand for skills and occupations based on historical data and economic trends. A case study mainly focused on predictive models for hiring time and wage range based on available job opening data. The research presents recommendations to HR professionals on how to predict the demand and prepare workers for future jobs. These recommendations will be useful in guiding workforce planning development strategies to ensure HR professionals are adequately prepared for the hiring process. INGARCH is discovered to perform the best in the prediction task's outcome, highest R-squared value of 0.37 and the lowest MAE and RMSE values of 18.46 and 36.05, respectively. The relatively poor performance of the ARIMA model may suggest that the job openings data may not be well described by a stationary process. The OLS model, the baseline model may be limited in capturing time series characteristics. However, all the forecasting models assessed in this thesis exhibited inadequate performance in explaining the variability in the data and making accurate predictions. This implies that the complexity of the data analyzed may surpass the capabilities of these models, or the algorithms utilized may require further refinement. Future research can concentrate on developing more effective forecasting models that can better accommodate complex data and improve predictive performance.Description
Thesis advisor
Malo, PekkaKeywords
workforce forecasting, demand prediction, time series data, machine learning