least squares

(redirected from Least-squares estimation)
Also found in: Dictionary, Thesaurus.

least squares

(lēst skwārz),
A principle of estimation invented by Gauss in which the estimates of a set of parameters in a statistical model are the quantities that minimize the sum of squared differences between the observed values of the dependent variable and the values predicted by the model.

least squares

a method of regression analysis. The line on a graph that best summarizes the relationship between two variables is the one that ensures that there is the least value of the sum of the squares of the deviation between the fitted curve and each of the original data points.
References in periodicals archive ?
Finally, examples of real geodetic networks will be given to illustrate those critical advantages of robust estimation versus least-squares estimation in case that undesired errors enter into the adjustment stage.
Therefore we will consider as the most suitable methodology for the geodetic network adjustment the joint use of classic least-squares estimation and robust estimation.
As known, the classic least squares adjustment provides the most likely solution whereas any robust estimator yields a sub-optimal solution close to the least-squares estimation.
The least-squares estimation may then be more or less sensitive to this error depending on its magnitude, network redundancy and the particular observation that is affected.
This second error will prove to be totally undetectable for the [tau]-test over the least-squares estimation and even for robust estimation performed by IRLS due to the fact it has little influence on the network inner consistence.
Regression 4 uses a static-score model that is estimated using a two-stage least-squares estimation with a fixed-and-period effects specification.

Full browser ?