linear regression

(redirected from regression line)
Also found in: Dictionary, Thesaurus, Financial, Encyclopedia, Wikipedia.
Related to regression line: regression analysis, correlation coefficient

linear regression

a statistical procedure in which a straight line is established through a data set that best represents a relationship between two subsets or two methods.

linear regression

A statistical method defined by the formula y = mx = b which is used to "best-fit" straight lines to scattered data points of paired values Xi, Yi, where the values of Y—the ordinate or vertical line—are “observations” or values of a variable (e.g., systolic blood pressure) and the values of X—the abscissa or horizontal line—increased in a relatively nonrandom fashion (e.g., age). Linear regression is a parametric procedure for determining the relationship between one or more (multiple) continuous or categorical predictor (or independent) variables and a continuous outcome (or dependent) variable.

In the equation y = mx = b:
m = slope
b = y - intercept

linear regression

Statistics A statistical method defined by the formula y = a + bx, which is used to 'fit' straight lines to scattered data points of paired values Xi, Yi, where the values of Y–the ordinate or vertical line are observations of a variable–eg, systolic BP and the values of X–the abscissa or horizontal line ↑ in a relatively nonrandom fashion–eg, age

linear regression

A statistical method of predicting the value of one variable, given the other, in a situation in which a CORRELATION is known to be significant. The equation is y = a + bx in which x and y are, respectively, the independent and dependent variables and a and b are constants. This is an equation for a straight line.


in a line.

linear assessment
a method of expressing an assessment result as a score out of a possible perfect score of 10, or some other number. Used in body condition scoring, showring judging of conformation.
linear dodecyl benzene sulfonic acid teat dip
linear energy transfer
expresses the quality of electronic radiation. It is concerned with the spatial distributions of energy transfers which occur in the tracks of particles as they penetrate matter.
linear program
a management program used to determine the best mix of ingredients or services to be used in a particular situation to maintain the highest level of productivity or profitability or other similar parameter.
linear regression
statistical method used to study the relationship between independent and dependent variables when the dependent variable consists of continuous data.
linear score
for somatic cell counts in milk (SCCs) convert SCC logarithmically from cells per milliliter to a linear score from 0-9. The linear score has a straight line, inverse relationship with milk yield. An increase of one in the linear score is associated with a 400-pound decrease in lactation milk yield or a 1.5 pound drop in daily yield.


1. return to a former or earlier state.
2. subsidence of clinical signs or of a disease process.
3. in biology, the tendency in successive generations toward the mean.
4. the relationship between pairs of random variables; the mean of one variable and its location is influenced by another variable.

regression analysis
see regression analysis.
regression coefficient
is the factor which determines the slope of a regression line; the greater the coefficient the steeper the line.
curvilinear regression
when the relationship between two variables is not linear.
linear regression
the relationship between two variables is a straight line.
References in periodicals archive ?
p], which is given by the slope of the regression lines, one sees that although there is some increase in this parameter with increasing thickness (more evident in SENT than in DENT specimens), the main difference is between the two geometries used with SENT, giving a consistently higher value than DENT specimens.
However, the regression line intercept for the first zone gave a significantly better correlation (r = 0.
The 1980 projection had the only regression line with coefficients consistent with the hypothesis of projected, as a group, being like the actual.
We calculated the linear regression equations of Cq vs RIN values and plotted the regression lines.
It is as if the data points are gradually included within, or excluded from, the band when the regression line is moving.
The solid vertical line next to the bracket represents the variation accounted for by the regression line (explained variation), while the dotted line represents what is not accounted for (unexplained variation, or error).
The slope and intercept of the fitted regression line at the fleet level (Fig.
6; the critical displacement is now located at the intersection between two linear regression lines.
With this template, the follow ing statistics are also generated: correlation coefficient, slope, yintercept, coefficient of determination, coefficient of non-deter mination, residual variance, tan dard error of the regression line, standard eff or of the slope, dard error of the y-intercept, and the 95 per cent confidence interval for the standard error of the last two statistics.
i in both of regression and artificial neural network methods, the rate of the deviation absolute value from the regression line from the actual yield (Yi) were calculated and compared and its comparison results using paired t test was listed in Table 2.