Prediction in statistical learning
Prediction problem is defined when we have a significant number of observations and their corresponding outcomes, and we'd like to predict the outcomes for another set of observations where the corresponding outcomes are unavailable. We call the set of all input variables , and the corresponding set of all outcomes . If we define as the estimate of the function that we need to estimate to map to , and we define as the resulting prediction of , and considering the fact that the error term averages to 0, the following holds:
0
5
Tags
Data Science
Related
Prediction in statistical learning
Inference in statistical learning
Prediction in statistical learning
10 themes for Superforecasting
A city's transportation department is considering adding a new bus line. Two planners present their forecasts for its potential ridership:
- Planner A: "My cousin started taking the bus last year and loves it. People are tired of driving. I predict this new line will be immediately popular and serve 5,000 riders per day within the first month."
- Planner B: "Surveys on the proposed route show high interest from 20% of residents. Similar lines in three comparable cities saw a 5-8% ridership increase in their first year. I predict the new line will serve 1,500-2,000 riders per day after six months of operation."
Based on the principles of making a sound forecast, which planner's prediction is more robust and why?
Evaluating a Business Sales Forecast
Match each statement to the category that best describes it. To do this, you must analyze whether the statement is about the future, if it is based on evidence, and if it can be objectively verified.