The main goal of this project is to leverage public available data sources (free and paid) to create a data science model and leverage the effects of a company's stock market earnings release on the company's stock value. In other words, how best to make money from earnings.

Every quarter publicly traded companies release their earnings reports and this affects the value of the stock. While this is a given, how do we determine which company stocks are most like to move and by how much.

• 
  
• Business Problem Definition: (1) Here we will properly identify and isolate the problem parameters (2) We will then craft a Vision, (3) Develop the KPIs, Timelines and and milestones - 1- 2 weeks
• Theoretical Formulation & Hypothesis: We will establish a sound theoretical test case or argument (i.e. English to numbers and math - 2 days - 1 week
• Modeling & computational Formalism - We will convert our theoretical test case to a computational framework. We want to understand how all our ideas translate from math to computation and the best computational approach and why 2- 3 weeks
• Data Gathering and Data integration : We have data sources available including data dumps (i.e. Static), , API, WebSocket, and streaming data but we need to step up our environment to ingest the data properly. 2- 3 weeks
• Data Manipulation and Processing (EDA, data validation) 1 week
• Software model building (Coding, Training and parameter tuning) 2 weeks.
• Testing (unit test, validation test) 1.5 weeks
• Model test (to know how accurate the model is and what the time drift is)
• Implementation and integration existing services (putting a wrapper and integrating with triggers and other software 2 weeks
• If time permits Execution of suggested approaches on an account)

As expected these generally are defined by

•  
• Analyzing the existing data sources and considering the data quality and availability 
• Developing a data science model 
• Consider and apply ways to clean the data if necessary, making it suitable for analysis 
• Optimizing model performance and assessing areas of improvement 
• Researching other variables that could enhance and improve the model. 
• Testing the developed model and making adjustments as necessary 
• Presenting initial findings on the data with an engaging data narrative 
• Develop initial interactive data visualizations that allow for the user to explore the data available 
•  

Our Bespoke software company develops many integrated data science projects which we go through the steps above for different clients. Our work is monitored with project management software and noted via emails. We have other projects in data science in Salesforce CRM and Rest estate as needed by clients

Access to all available data and systems, mentorship time with our engineering team, support with key metrics and performance indicators, and overall supervision