Chapter 6. Forecasting Numeric Data – Regression Methods
Mathematical relationships describe many aspects of everyday life. For example, a person's body weight can be described in terms of his or her calorie intake; one's income can be related to years of education and job experience; and the president's odds of being re-elected can be estimated by popular opinion poll numbers.
In each of these cases, numbers specify precisely how the data elements are related. An additional 250 kilocalories consumed daily is likely to result in nearly a kilogram of weight gain per month. Each year of job experience may be worth an additional $1,000 in yearly salary while years of education might be worth $2,500. A president is more likely to be re-elected with a high approval rating. Obviously, these types of equations do not perfectly model every case, but on average, the rules might work fairly well.
A large body of work in the field of statistics describes techniques for estimating such numeric relationships...