least squares
1
7.0
- noun
- /liːst skwɛrz/
- Specialized
:
mínimos cuadrados
- In this study, we applied the least squares method to find the best-fitting line for our data.
- ordinary least squares
- least squares regression
- weighted least squares
But but so this least squares is is not deep learning.
- But but so this least squares is is not deep learning.
- In these situations, a least square Monte Carlo method with a backward induction approach is used.
Examples
-
Once the measurement has been made, one can use a least-squares or similar technique to estimate the model parameters.
Academic text (1992) -
An ordinary least squares regression was conducted for each of the six dependent variables.
Academic text (2013) -
OK, well, the least squares fit is this dashed line.
-
Then we got, you know, ordinary least squares linear regression.
-
Remember that the formula for the least squares line is Y equals AX plus B.
-
The weighted least squares mean d is the measure of interest when conducting the post-hoc analysis.
Academic text (1999) -
For all analyses, we used robust least squares estimation procedures.
Academic text (2008) -
I estimated ordinary least squares (OLS) models of obesity and BMI by using the main sample.
Academic text (2011) -
I estimated ordinary least squares models of obesity, body mass index (BMI), and change in BMI.
Academic text (2011) -
To examine these predictors of associational membership, an ordinary least squares (OLS) regression analysis was used (see appendix).
Academic text (1999)
Synonyms
A way to fit a line or curve to data by making squared errors as small as possible
Surface Forms
Morphology
The compound literally links 'least' (smallest) with 'squares' (squared numbers), so a learner who knows both words can infer it involves making squared values as small as possible — a partially compositional relation. However, the specific statistical procedure (minimizing the sum of squared deviations to fit a curve) is a specialised mathematical concept that a general B1 learner would not fully derive from the words alone, so the expression is only partially transparent.
Etymology
Least squares comes from a simple idea: change each measurement 'error' into a square by multiplying the number by itself, and then pick the line or curve that has the smallest total of those squares. So least means the smallest total, and the name explains how the method finds the 'best fit' to the data.