kstats-correlation/org.oremif.kstats.correlation/SimpleLinearRegressionResult

SimpleLinearRegressionResult

class SimpleLinearRegressionResult(source)

The result of a simple linear regression fit (y = intercept + slope * x).

Example:

val result = simpleLinearRegression(x, y)
result.slope     // change in y per unit change in x
result.intercept // predicted y when x = 0
result.rSquared  // proportion of variance explained (0.0 to 1.0)
result.predict(6.0) // predict y for a new x value

Properties

intercept

val intercept: Double

the predicted value of y when x is zero.

val n: Int

the number of observations used in the regression.

residuals

val residuals: DoubleArray

The difference between each observed y value and the predicted value (y - predicted). Residuals sum to approximately zero for a correctly fitted model.

rSquared

val rSquared: Double

the coefficient of determination, indicating the proportion of variance in y explained by the linear relationship with x. Ranges from 0.0 (no explanatory power) to 1.0 (perfect fit). Returns 1.0 when y has zero variance (all values identical), since there is no variance left to explain.

slope

val slope: Double

the change in y for a one-unit increase in x.

standardErrorIntercept

val standardErrorIntercept: Double

the standard error of the intercept estimate, measuring the uncertainty in intercept.

standardErrorSlope

val standardErrorSlope: Double

the standard error of the slope estimate, measuring the uncertainty in slope. Smaller values indicate a more precise estimate.