The term "R-squared," or the coefficient of determination, explains the percent of variance away from a dependent variable and is expressed as a percentage between 0 and 100. An R-squared value explains how data fits a statistical set of numbers, sometimes expressed on a graph as a line or curve surrounded by points. The closer the R-squared value is to 100, the more dependent that value is on another variable.
Continue ReadingR-squared measures an investment security's performance based on a benchmark index for that particular type of security. With investing, a number closer to 100 delineates a security completely dependent on an index. For example, a stock with an R-squared value of 100 has a price that is totally dependent on the index's daily movements. An R-squared number of 0 for a security means the value is not determined by an index whatsoever. Fixed-income securities are indexed by T-bill rates. Equity values are based on the benchmark S&P 500.
Statistics models do not necessarily have a "good" or "bad" range for R-squared, because the figure depends on the type of variable measured against it. A low R-squared value may be a good indication over time. However, a high R-squared number may indicate good news when the figure compares raw statistics in a stationary series that does not depend on chronological factors.
Learn more about Statistics