Key Terms and Theories PDF

TOPIC 1 Key Terms and Theories 1. Business Analytics Descriptive Analytics: Extracting insights from historical data using techniques such as data visualization. Predictive Analytics: Using statistical models to predict future outcomes....

TOPIC 1 Key Terms and Theories 1. Business Analytics Descriptive Analytics: Extracting insights from historical data using techniques such as data visualization. Predictive Analytics: Using statistical models to predict future outcomes. Prescriptive Analytics: Recommending decisions based on data and optimization models. 2. Management Science Focuses on applying scientific approaches to solve managerial problems. Originates from operations research** used in military operations during WWII. 3. Optimization Models Objective Function: The function that needs to be maximized or minimized (e.g., profit). Decision Variables**: Variables whose values are determined to optimize the objective. Constraints: Restrictions applied to decision variables (e.g., budget limits). 4. Decision Analysis Maximax Criterion: For optimists, focuses on the maximum possible payoff. Maximin Criterion: For pessimists, focuses on minimizing the worst possible outcome. Bayes' Decision Rule: Incorporates prior probabilities to make decisions that maximize expected payoff. 5. Data Science An interdisciplinary field that extracts knowledge from large datasets using scientific methods and algorithms. 6. Machine Learning A subset of AI, it allows computers to learn from data to make predictions and decisions automatically. 7. Artificial Intelligence Simulates human intelligence in machines; includes machine learning, robotics, natural language processing, etc. 8. Regression Analysis Linear Regression: A method for modeling the relationship between a dependent variable and one or more independent variables. Polynomial Regression: Used for modeling more complex relationships between variables. 9. Utility Theory Deals with risk preferences in decision making. Individuals may choose a guaranteed outcome over a probabilistic one with a higher expected payoff due to risk aversion. 10. Data Dictionary Defines variables in datasets, specifying data formats and units (e.g., credit score, loan size, interest rate). 11. Decision Trees A graphical decision-making tool that helps to identify the best course of action based on probabilities of different outcomes. 12. Heuristics and Metaheuristics Heuristics: Simple, intuitive rules used to make decisions without necessarily finding an optimal solution. Metaheuristics: More advanced algorithms used to solve complex optimization problems by providing guidelines for designing heuristics. 13. Post-Optimality Analysis A process of evaluating how changes in parameters would affect the optimal solution of a decision model. This is often referred to as **sensitivity analysis**. 14. Expected Value of Perfect Information (EVPI) Represents the difference between the expected payoff with perfect information and without perfect information, indicating how much value additional information could provide. 15. Bayesian Updating The process of adjusting probabilities based on new data or evidence, especially relevant in decision-making under uncertainty. 16. Payoff Table A table summarizing the possible payoffs for each decision alternative under different states of nature. 17. Constraints Restrictions or limitations applied to decision variables (e.g., budget, time, or resource limits). 18. Objective Function The mathematical expression representing the goal (such as profit) that needs to be optimized. 19. Maximax and Maximin Criteria Maximax: Focuses on the most optimistic outcome, maximizing the best-case scenario. Maximin: Focuses on minimizing the worst-case scenario, more conservative decision-making. 20. Overfitting In predictive analytics, it refers to a model that fits historical data too closely, capturing noise instead of the underlying pattern, which leads to poor performance on new data. 21. Sensitivity Analysis Evaluating how sensitive a solution is to changes in input parameters or assumptions.