statistical procedure that attempts to assess the relationship between a dependent variable andtwoor more independent variables. Examples: Total factory overhead (the dependent variable) is related to both labor-hours and machinehours (the independent variables). Sales of a popular soft drink (the dependent variable) is a function of various factors, such as its price, advertising, taste, and the prices of its major competitors (the independent variables).

condition that exists when independent variables are highly correlated with each other. In the presence of multicollinearity, the estimated regression coefficient may be unreliable. The presence of multicollinearity can be tested by investigating the correlation (r) between the independent variables.

typically used by tax assessors, an effort to determine salient characteristics of properties in a given submarket, to allow an approximation of value for each. Sophisticated statistical techniques are used frequently in mass appraising.

Example: Through use of multiple regression statistics, the tax assessor developed the formula in Table 31 to estimate the value of homes in a certain neighborhood.

A2,000-square-foot home, 10 years old, with one fireplace, a garage, and two baths would be assessed at $127,000 as follows:

a statistical technique used to estimate mathematical models of economic and other processes. It is used to find a mathematical expression that best fits the relationship between a group of random variables as indicated by a sample of data.

Example: An appraiser wants to find the relationship between sales prices for homes and their physical characteristics. Data are collected on the prices of a group of homes and their size, number of rooms, location, and age. Linear regression is used to analyze the relationship expressed by the data. The regression model can then be used to estimate prices for other similar houses on the market.