EPS Forecast Data Analysis: EPS Prediction Model Development Overview In this project, you will develop a predictive model for Earnings Per Share (EPS) for one hypothetical client Sink or Swim (SOS) using analysts' forecast data provided by Wharton Research Data Services (WRDS) via the Institutional Brokers' Estimate System (I/B/E/S). This project focuses solely on constructing a model to predict EPS, a crucial financial indicator for both our hypothetical clients and their decision-making processes. Objectives Conduct data cleaning and exploratory data analysis Develop a predictive model for EPS using historical data. Implement feature engineering to improve model accuracy. Apply and compare at least four different modeling techniques. Conduct model selection using best subset, forward, and backward selection methods, coupled with cross-validation. Data Dictionary: You will work with analysts' forecast data of earnings per share (EPS) provided by Wharton Research Data Services (WRDS). Institutional Brokers' Estimate System (I/B/E/S) provides historical data on certain financial indicators collected from thousands of individual analysts working in more than 3,000 broker houses. TICKER: A unique identifier assigned to each security. CNAME: Company name ACTDATS: The Activation date: It is the date when the analyst forecast became effective within the IBES database. ESTIMATOR: Sellside institution (mostly broker house). It is just the broker. ANALYS: The person