SuanShu, a Java numerical and statistical library
A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

R

R() - Method in class com.numericalmethod.suanshu.matrix.doubles.factorization.qr.GramSchmidt
 
R() - Method in class com.numericalmethod.suanshu.matrix.doubles.factorization.qr.HouseholderReflection
 
R() - Method in class com.numericalmethod.suanshu.matrix.doubles.factorization.qr.QR
 
R() - Method in interface com.numericalmethod.suanshu.matrix.doubles.factorization.qr.QRDecomposition
Get a copy of the upper triangular matrix R in the QR decomposition.
R - Class in com.numericalmethod.suanshu.misc
This class collects some R-equivalent utility functions.
r - Variable in class com.numericalmethod.suanshu.optimization.constrained.linearprogramming.simplex.pivoting.SimplexPivoting.Pivot
pivot row
r(CointegrationMle, double) - Method in class com.numericalmethod.suanshu.stats.cointegration.JohansenTest
Get the (most likely) order of cointegration.
r - Variable in class com.numericalmethod.suanshu.stats.random.pseudorandom.linear.Lehmer
m % a
R.ifelse - Interface in com.numericalmethod.suanshu.misc
the placeholder for R.ifelse
R.which - Interface in com.numericalmethod.suanshu.misc
the placeholder for R.which
R1toConstantMatrix - Class in com.numericalmethod.suanshu.analysis.function.matrix
This class represents a matrix function that always returns the same constant matrix.
R1toConstantMatrix(Matrix) - Constructor for class com.numericalmethod.suanshu.analysis.function.matrix.R1toConstantMatrix
Construct a constant matrix function.
R1toMatrix - Class in com.numericalmethod.suanshu.analysis.function.matrix
This abstract class represents a R2 → Matrix space function.
R1toMatrix() - Constructor for class com.numericalmethod.suanshu.analysis.function.matrix.R1toMatrix
 
R2 - Variable in class com.numericalmethod.suanshu.stats.regression.linear.ols.Residuals
diagnostic measure: the R-squared
R2toMatrix - Class in com.numericalmethod.suanshu.analysis.function.matrix
This abstract class represents a R2 → Matrix space function.
R2toMatrix() - Constructor for class com.numericalmethod.suanshu.analysis.function.matrix.R2toMatrix
 
random() - Static method in class com.numericalmethod.suanshu.stats.random.RNG
Get a double between 0 and 1.
random - Variable in class com.numericalmethod.suanshu.stats.timeseries.linear.univariate.stationaryprocess.MADecomposition
the stationary random component of the time series after the trend and seasonal components are removed
RandomLongGenerator - Interface in com.numericalmethod.suanshu.stats.random
A (pseudo) random number generator that generates a sequence of longs that lack any pattern.
RandomNumberGenerator - Interface in com.numericalmethod.suanshu.stats.random
A (pseudo) random number generator is an algorithm designed to generate a sequence of numbers that lack any pattern.
RandomProcess - Interface in com.numericalmethod.suanshu.stats.stochasticprocess
This interface represents a random process a.k.a. stochastic process.
RandomVectorGenerator - Interface in com.numericalmethod.suanshu.stats.random.multivariate
A (pseudo) multivariate random number generator samples a random vector from a multivariate distribution.
RandomWalk - Class in com.numericalmethod.suanshu.stats.stochasticprocess.multivariate.brownian
This is the Random Walk construction of a multivariate Brownian motion.
RandomWalk(int, TimeGrid) - Constructor for class com.numericalmethod.suanshu.stats.stochasticprocess.multivariate.brownian.RandomWalk
Construct a multi-dimensional Brownian motion at time points specified.
RandomWalk(int, int) - Constructor for class com.numericalmethod.suanshu.stats.stochasticprocess.multivariate.brownian.RandomWalk
Construct a multi-dimensional Brownian motion at even time points, [0, 1, ......, T].
RandomWalk - Class in com.numericalmethod.suanshu.stats.stochasticprocess.multivariate.integration.sde
This is the Random Walk construction of a stochastic process per SDE specification.
RandomWalk(DiscretizedSDE, TimeGrid) - Constructor for class com.numericalmethod.suanshu.stats.stochasticprocess.multivariate.integration.sde.RandomWalk
Construct a multivariate stochastic process from an SDE.
RandomWalk - Class in com.numericalmethod.suanshu.stats.stochasticprocess.univariate.brownian
This is the Random Walk construction of a univariate Brownian motion.
RandomWalk(TimeGrid) - Constructor for class com.numericalmethod.suanshu.stats.stochasticprocess.univariate.brownian.RandomWalk
Construct a univariate Brownian motion at time points specified.
RandomWalk(int) - Constructor for class com.numericalmethod.suanshu.stats.stochasticprocess.univariate.brownian.RandomWalk
Construct a univariate Brownian motion at even time points, [0, 1, ......, T].
RandomWalk - Class in com.numericalmethod.suanshu.stats.stochasticprocess.univariate.integration.sde
This is the Random Walk construction of a stochastic process per SDE specification.
RandomWalk(DiscretizedSDE, TimeGrid) - Constructor for class com.numericalmethod.suanshu.stats.stochasticprocess.univariate.integration.sde.RandomWalk
Construct a univariate stochastic process from an SDE.
RandomWalk.Iterator - Class in com.numericalmethod.suanshu.stats.stochasticprocess.multivariate.integration.sde
This iterates a deterministic realization (time series) of a multivariate stochastic process.
RandomWalk.Iterator - Class in com.numericalmethod.suanshu.stats.stochasticprocess.univariate.integration.sde
This iterates a deterministic realization (time series) of a univariate stochastic process.
RandomWalk.MultiVariateRealization - Class in com.numericalmethod.suanshu.stats.stochasticprocess.multivariate.integration.sde
 
RandomWalk.Realization - Class in com.numericalmethod.suanshu.stats.stochasticprocess.univariate.integration.sde
 
rank() - Method in class com.numericalmethod.suanshu.matrix.doubles.factorization.qr.GramSchmidt
 
rank() - Method in class com.numericalmethod.suanshu.matrix.doubles.factorization.qr.HouseholderReflection
Rank is computed by counting the number of non-zero rows in R.
rank() - Method in class com.numericalmethod.suanshu.matrix.doubles.factorization.qr.QR
 
rank() - Method in interface com.numericalmethod.suanshu.matrix.doubles.factorization.qr.QRDecomposition
Get the numerical rank of the matrix A as computed by the QR decomposition.
rank() - Method in class com.numericalmethod.suanshu.matrix.doubles.linearsystem.Kernel
Get the rank of A.
rank(Matrix, double) - Static method in class com.numericalmethod.suanshu.matrix.doubles.operation.Measure
Compute the numerical rank of a matrix.
rank(Matrix) - Static method in class com.numericalmethod.suanshu.matrix.doubles.operation.Measure
Compute the numerical rank of a matrix.
Rank - Class in com.numericalmethod.suanshu.stats.descriptive.rank
Rankin is a relationship between a set of items such that, for any two items, the first is either "ranked higher than", "ranked lower than" or "ranked equal to" the second.
Rank(double[], double) - Constructor for class com.numericalmethod.suanshu.stats.descriptive.rank.Rank
Compute the sample ranks of the values.
Rank(double[]) - Constructor for class com.numericalmethod.suanshu.stats.descriptive.rank.Rank
Compute the sample ranks of the values using the machine epsilon as the tie threshold.
rank(int) - Method in class com.numericalmethod.suanshu.stats.descriptive.rank.Rank
Get the rank of the element arr[i].
rank() - Method in class com.numericalmethod.suanshu.vector.doubles.dense.operation.VectorSpace
Get the rank of this vector space.
RankOne - Class in com.numericalmethod.suanshu.optimization.unconstrained.quasinewton
The Rank One method is a quasi-Newton method to solve unconstrained nonlinear optimization problems.
RankOne() - Constructor for class com.numericalmethod.suanshu.optimization.unconstrained.quasinewton.RankOne
Construct an instance of RankOne to minimize a function.
ranks() - Method in class com.numericalmethod.suanshu.stats.descriptive.rank.Rank
Get a copy of the ranks.
rate - Variable in class com.numericalmethod.suanshu.analysis.differentiation.Ridders
the rate at which h decreases
rate - Variable in class com.numericalmethod.suanshu.analysis.integration.univariate.riemann.EulerMaclaurin
the rate of sub-dividing an interval Starting with the whole interval (b-a), we divide each sub-interval into rate many intervals.
ratioTest(Tableau, int) - Method in class com.numericalmethod.suanshu.optimization.constrained.linearprogramming.simplex.pivoting.NaiveRule
pivot row selection (Ratio test): The pivot row is the smallest basic variable index, r, such that row r satisfies -h(r) / H(r,s) = min {-h(i) / H(i,s) | H(i,s) < 0}
ratioTest(Tableau, int) - Method in interface com.numericalmethod.suanshu.optimization.constrained.linearprogramming.simplex.pivoting.SimplexPivoting
pivot row selection (Ratio test)
Rayleigh - Class in com.numericalmethod.suanshu.stats.random.distribution
Sample pseudo random numbers from the RayleighDistribution distribution.
Rayleigh(double, RandomNumberGenerator) - Constructor for class com.numericalmethod.suanshu.stats.random.distribution.Rayleigh
Construct a pseudo-random number generator of a RayleighDistribution distribution, using a customized uniform random number generator.
Rayleigh(double) - Constructor for class com.numericalmethod.suanshu.stats.random.distribution.Rayleigh
Construct a pseudo-random number generator of a RayleighDistribution distribution.
RayleighDistribution - Class in com.numericalmethod.suanshu.stats.distribution.univariate
The L2 norm of (x1, x2), where xi's are normal, uncorrelated, equal variance and have RayleighDistribution distributions.
RayleighDistribution(double) - Constructor for class com.numericalmethod.suanshu.stats.distribution.univariate.RayleighDistribution
Construct a RayleighDistribution distribution.
rbind(Vector...) - Static method in class com.numericalmethod.suanshu.matrix.doubles.operation.CreateMatrix
Take an array of vectors and combine them by rows.
rbind(VectorList) - Static method in class com.numericalmethod.suanshu.matrix.doubles.operation.CreateMatrix
Take a list of vectors and combine them by rows.
rbind(Matrix...) - Static method in class com.numericalmethod.suanshu.matrix.doubles.operation.CreateMatrix
Take an array of matrices and combine them by rows.
rbind(MatrixList) - Static method in class com.numericalmethod.suanshu.matrix.doubles.operation.CreateMatrix
Take a list of matrices and combine them by rows.
real - Variable in class com.numericalmethod.suanshu.number.complex.Complex
the real part of this complex number
Real - Class in com.numericalmethod.suanshu.number
This class represents an arbitrary precision real number.
Real(double) - Constructor for class com.numericalmethod.suanshu.number.Real
Construct a Real from a double.
Real(long) - Constructor for class com.numericalmethod.suanshu.number.Real
Construct a Real from an integer.
Real(BigDecimal) - Constructor for class com.numericalmethod.suanshu.number.Real
Construct a Real from a BigDecimal.
Real(BigInteger) - Constructor for class com.numericalmethod.suanshu.number.Real
Construct a Real from a BigInteger.
Real(String) - Constructor for class com.numericalmethod.suanshu.number.Real
Construct a Real from a String representation of a number.
realEigenvalues() - Method in class com.numericalmethod.suanshu.matrix.doubles.factorization.eigen.Eigen
Get the real eigenvalue.
realization(Vector) - Method in class com.numericalmethod.suanshu.stats.stochasticprocess.multivariate.brownian.RandomWalk
 
realization(Vector) - Method in interface com.numericalmethod.suanshu.stats.stochasticprocess.multivariate.integration.sde.Construction
Construct a realization of a stochastic process.
realization(Vector) - Method in class com.numericalmethod.suanshu.stats.stochasticprocess.multivariate.integration.sde.RandomWalk
 
realization(double) - Method in class com.numericalmethod.suanshu.stats.stochasticprocess.univariate.brownian.RandomWalk
 
realization(double) - Method in interface com.numericalmethod.suanshu.stats.stochasticprocess.univariate.integration.sde.Construction
Construct a realization of the stochastic process.
realization(double) - Method in class com.numericalmethod.suanshu.stats.stochasticprocess.univariate.integration.sde.RandomWalk
 
Realization - Interface in com.numericalmethod.suanshu.stats.stochasticprocess.univariate
This interface defines the Iterator for generating (reading) a realization of a univariate random process.
Realization - Interface in com.numericalmethod.suanshu.stats.timeseries.univariate.realtime
This represents a univariate time series indexed real numbers.
Realization.Entry - Class in com.numericalmethod.suanshu.stats.timeseries.univariate.realtime
the TimeSeries.Entry for a real number -indexed univariate time series
Realization.Entry(double, double) - Constructor for class com.numericalmethod.suanshu.stats.timeseries.univariate.realtime.Realization.Entry
 
Realization.Iterator - Class in com.numericalmethod.suanshu.stats.stochasticprocess.univariate
This Iterator support lazy evaluation/generation of a realization from a stochastic process.
Realization.Iterator(Realization, long) - Constructor for class com.numericalmethod.suanshu.stats.stochasticprocess.univariate.Realization.Iterator
Construct a realization of a univariate stochastic process.
Realization.Iterator - Class in com.numericalmethod.suanshu.stats.timeseries.univariate.realtime
the Iterator to read a real number -indexed univariate time series
Realization.Iterator(Realization) - Constructor for class com.numericalmethod.suanshu.stats.timeseries.univariate.realtime.Realization.Iterator
 
RealMatrix - Class in com.numericalmethod.suanshu.matrix.generic.matrixtype
This class represents a matrix of Real numbers.
RealMatrix(int, int) - Constructor for class com.numericalmethod.suanshu.matrix.generic.matrixtype.RealMatrix
Construct a nRows x nCols matrix of Real numbers.
RealMatrix(Real[][]) - Constructor for class com.numericalmethod.suanshu.matrix.generic.matrixtype.RealMatrix
Construct a matrix from a 2D array of Real numbers.
RealMatrix(double[][]) - Constructor for class com.numericalmethod.suanshu.matrix.generic.matrixtype.RealMatrix
Construct a RealMatrix from a 2D array of doubles.
realPart(double[]) - Method in class com.numericalmethod.suanshu.analysis.function.rn2r1.RealScalarFunctionFixedVariables
Given a vector input to the original function, this extracts the real parts (excluding the fixed integer values).
realRoots() - Method in class com.numericalmethod.suanshu.analysis.function.polynomial.root.Roots
Get a copy of all the real roots of the polynomial.
RealScalarFunction - Interface in com.numericalmethod.suanshu.analysis.function.rn2r1
This interface represents a Rn → R1 function, y = f(x0, x1, ..., xn) The function takes n real arguments and output one real value.
RealScalarFunctionFixedVariables - Class in com.numericalmethod.suanshu.analysis.function.rn2r1
This creates a a RealScalarFunction from another RealScalarFunction by fixing the values of a subset of variables.
RealScalarFunctionFixedVariables(RealScalarFunction, Collection<RealScalarFunctionFixedVariables.Value>) - Constructor for class com.numericalmethod.suanshu.analysis.function.rn2r1.RealScalarFunctionFixedVariables
 
RealScalarFunctionFixedVariables(RealScalarFunction, RealScalarFunctionFixedVariables.Value...) - Constructor for class com.numericalmethod.suanshu.analysis.function.rn2r1.RealScalarFunctionFixedVariables
 
RealScalarFunctionFixedVariables.Value - Class in com.numericalmethod.suanshu.analysis.function.rn2r1
 
RealScalarFunctionFixedVariables.Value(int, double) - Constructor for class com.numericalmethod.suanshu.analysis.function.rn2r1.RealScalarFunctionFixedVariables.Value
 
RealVectorFunction - Interface in com.numericalmethod.suanshu.analysis.function.rn2rm
This interface represents a Rn → Rm function, [y0, y1, y2, ...] = f(x0, x1, ..., xn) The function takes n real arguments and output m real values in the form of a Vector.
reason() - Method in exception com.numericalmethod.suanshu.matrix.doubles.matrixtype.sparse.solver.iterative.IterativeSolver.ConvergenceFailure
Get the reason for the convergence failure.
reduce(Matrix) - Method in class com.numericalmethod.suanshu.matrix.doubles.factorization.eigen.Hessenberg
Deprecated. Not supported yet.
reducedRowEchelonForm(Matrix, double) - Static method in class com.numericalmethod.suanshu.matrix.doubles.IsMatrix
Check if a matrix is in the reduced row echelon form.
Reference<T> - Class in com.numericalmethod.suanshu.parallel
 
Reference() - Constructor for class com.numericalmethod.suanshu.parallel.Reference
 
reflect(Vector) - Method in class com.numericalmethod.suanshu.matrix.doubles.operation.Householder
Apply the Householder matrix, H, to a column vector, x.
reflect(Matrix) - Method in class com.numericalmethod.suanshu.matrix.doubles.operation.Householder
Apply the Householder matrix, H, to a matrix (a set of column vectors), A.
reflectRows(Matrix) - Method in class com.numericalmethod.suanshu.matrix.doubles.operation.Householder
Apply the Householder matrix, H, to a matrix (a set of row vectors), A.
rejectNull(double) - Method in class com.numericalmethod.suanshu.stats.test.HypothesisTest
Use p-value to check whether the null hypothesis can be rejected for given significance level (size) alpha.
relations(Interval<T>) - Method in class com.numericalmethod.suanshu.interval.Interval
Determine the relations between this and Y.
relativeError(double, double) - Static method in class com.numericalmethod.suanshu.number.DoubleUtils
Compute the relative error for {x1, x0}.
RelativeTolerance - Class in com.numericalmethod.suanshu.matrix.doubles.matrixtype.sparse.solver.iterative
The stopping criteria is that the norm of the residual r relative to the input base is equal to or smaller than the specified tolerance, that is, ||r||2 ------ ≤ tolerance base For example, a Tolerance instance which compares the residual norm relative to the right-hand side b of a linear system Ax = b can be created as follows: Tolerance tolerance = new RelativeTolerance(b.norm()); Or, one which compares the residual norm relative to the initial residual can be created as follows: Tolerance tolerance = new RelativeTolerance(b.minus(A.multiply(x0)).norm());
RelativeTolerance(double) - Constructor for class com.numericalmethod.suanshu.matrix.doubles.matrixtype.sparse.solver.iterative.RelativeTolerance
Create an instance which uses RelativeTolerance.DEFAULT_TOLERANCE.
RelativeTolerance(double, double) - Constructor for class com.numericalmethod.suanshu.matrix.doubles.matrixtype.sparse.solver.iterative.RelativeTolerance
Create an instance which uses the specified tolerance.
remainder() - Method in class com.numericalmethod.suanshu.analysis.function.polynomial.HornerScheme
Get the remainder.
remove() - Method in class com.numericalmethod.suanshu.matrix.doubles.matrixtype.sparse.SparseVector.Iterator
Overriden to avoid the vector being altered.
remove() - Method in class com.numericalmethod.suanshu.stats.timeseries.TimeSeries.Iterator
Deprecated. a time series is immutable
rep(double, int) - Static method in class com.numericalmethod.suanshu.misc.R
This generates an array of doubles of repeated values.
rep(int, int) - Static method in class com.numericalmethod.suanshu.misc.R
This generates an array of ints of repeated values.
replace(Matrix, int, int, int, int, Matrix) - Static method in class com.numericalmethod.suanshu.matrix.doubles.operation.CreateMatrix
Replace a part of a matrix with a smaller matrix.
reset() - Method in class com.numericalmethod.suanshu.optimization.unconstrained.quasinewton.QuasiNewton.QuasiNewtonImpl
 
reset() - Method in class com.numericalmethod.suanshu.optimization.unconstrained.steepestdescent.SteepestDescent.LineSearch
Reset the transient states of the instance.
residuals - Variable in class com.numericalmethod.suanshu.stats.regression.linear.glm.GeneralizedLinearModel
the residual analysis of this GLM regression
residuals - Variable in class com.numericalmethod.suanshu.stats.regression.linear.glm.quasi.GeneralizedLinearModelQuasiFamily
the residual analysis of this quasi GLM regression
Residuals - Class in com.numericalmethod.suanshu.stats.regression.linear.glm.quasi
Residual analysis of the results of a quasi Generalized Linear Model regression.
Residuals - Class in com.numericalmethod.suanshu.stats.regression.linear.glm
Residual analysis of the results of a Generalized Linear Model regression.
Residuals(GlmProblem, Vector) - Constructor for class com.numericalmethod.suanshu.stats.regression.linear.glm.Residuals
Perform the residual analysis for a GLM problem.
residuals - Variable in class com.numericalmethod.suanshu.stats.regression.linear.logistic.Logistic
the residual analysis of this regression
Residuals - Class in com.numericalmethod.suanshu.stats.regression.linear.logistic
Residual analysis of the results of a logistic regression.
residuals - Variable in class com.numericalmethod.suanshu.stats.regression.linear.ols.OlsRegression
the residual analysis of this OLS regression
Residuals - Class in com.numericalmethod.suanshu.stats.regression.linear.ols
Residual analysis of the results of an Ordinary Least Square linear regression model.
Residuals - Class in com.numericalmethod.suanshu.stats.regression.linear
The residual of a sample is the difference between the sample and the estimated function (fitted) value.
Residuals(LmProblem, Vector) - Constructor for class com.numericalmethod.suanshu.stats.regression.linear.Residuals
Create an instance of Residuals for a linear regression problem.
residuals - Variable in class com.numericalmethod.suanshu.stats.regression.linear.Residuals
the residuals, ε
results() - Method in exception com.numericalmethod.suanshu.parallel.MultipleExecutionException
 
reverse(double...) - Static method in class com.numericalmethod.suanshu.number.DoubleUtils
Reverse a double[] array.
reverse(int...) - Static method in class com.numericalmethod.suanshu.number.DoubleUtils
Reverse an int[] array.
rho() - Method in class com.numericalmethod.suanshu.matrix.doubles.matrixtype.GivensMatrix
Get ρ as in Section 5.1.11 in Matrix Computations, 3rd edition, by Golub G.
rho - Variable in class com.numericalmethod.suanshu.optimization.unconstrained.linesearch.Fletcher
the precision parameter rho The smaller it is, e.g., 0.1, the more accurate the result is.
rho - Variable in class com.numericalmethod.suanshu.optimization.unconstrained.NelderMead
the contraction coefficient
Ridders - Class in com.numericalmethod.suanshu.analysis.differentiation
This class computes the numerical derivative of a function using Ridders' method.
Ridders(UnivariateRealFunction, int, double, int) - Constructor for class com.numericalmethod.suanshu.analysis.differentiation.Ridders
Construct the derivative function of a univariate function using Ridder's method.
Ridders(UnivariateRealFunction, int) - Constructor for class com.numericalmethod.suanshu.analysis.differentiation.Ridders
Construct the derivative function of a univariate function using Ridder's method.
Ridders(RealScalarFunction, int[], double, int) - Constructor for class com.numericalmethod.suanshu.analysis.differentiation.Ridders
Construct the derivative function of a real vector-valued function using Ridder's method.
Ridders(RealScalarFunction, int[]) - Constructor for class com.numericalmethod.suanshu.analysis.differentiation.Ridders
Construct the derivative function of a real vector-valued function using Ridder's method.
Riemann - Class in com.numericalmethod.suanshu.analysis.integration.univariate.riemann
This class uses an integrator together with Romberg's method.
Riemann() - Constructor for class com.numericalmethod.suanshu.analysis.integration.univariate.riemann.Riemann
Construct a default integrator.
Riemann(double, int) - Constructor for class com.numericalmethod.suanshu.analysis.integration.univariate.riemann.Riemann
Construct an integrator.
rightConfidenceInterval(double) - Method in class com.numericalmethod.suanshu.stats.test.mean.T
Compute the one sided right confidence interval, [a, ∞)
rightConfidenceInterval(double) - Method in class com.numericalmethod.suanshu.stats.test.variance.F
Compute the one sided right confidence interval, [a, ∞)
rightMultiply(Matrix) - Method in class com.numericalmethod.suanshu.matrix.doubles.matrixtype.GivensMatrix
Right multiplication by G, namely, A %*% G affects only the i-th and the j-th columns.
rightMultiply(Matrix) - Method in class com.numericalmethod.suanshu.matrix.doubles.matrixtype.PermutationMatrix
Right multiplication by P.
rightPreconditioner(Preconditioner) - Method in class com.numericalmethod.suanshu.matrix.doubles.matrixtype.sparse.solver.iterative.IterativeSolver.Problem
Override the right preconditioner.
rightPreconditioner() - Method in class com.numericalmethod.suanshu.matrix.doubles.matrixtype.sparse.solver.iterative.IterativeSolver.Problem
Get the right preconditioner.
rightTailApproximation - Variable in class com.numericalmethod.suanshu.stats.test.distribution.kolmogorov.KolmogorovDistribution
true if we use approximation for the right tail to speed up computation; up to 7 digit of accuracy
Ring<R> - Interface in com.numericalmethod.suanshu.mathstructure
This interface represents a ring structure.
RNG - Class in com.numericalmethod.suanshu.stats.random
This is the global pseudo-random number generator for the SuanShu library.
RNG() - Constructor for class com.numericalmethod.suanshu.stats.random.RNG
 
RntoMatrix - Interface in com.numericalmethod.suanshu.analysis.function.matrix
This interface represents a Rn → Matrix space function.
Romberg - Class in com.numericalmethod.suanshu.analysis.integration.univariate.riemann
Romberg's method computes an integral by generating a sequence of estimations of the integral values and then extrapolation.
Romberg(IterativeIntegrator) - Constructor for class com.numericalmethod.suanshu.analysis.integration.univariate.riemann.Romberg
Construct an integrator by extending an IterativeIntegrator using Romberg's method.
ROOT_2 - Static variable in class com.numericalmethod.suanshu.Constant
√2
ROOT_2_PI - Static variable in class com.numericalmethod.suanshu.Constant
√(2π)
ROOT_PI - Static variable in class com.numericalmethod.suanshu.Constant
√π
Roots - Class in com.numericalmethod.suanshu.analysis.function.polynomial.root
This class provides utility functions to access the roots returned by Solver.
Roots(NumberList) - Constructor for class com.numericalmethod.suanshu.analysis.function.polynomial.root.Roots
Construct a Roots instance to collect all the roots of a polynomial equation.
rotate(Vector) - Method in class com.numericalmethod.suanshu.matrix.doubles.matrixtype.GivensMatrix
Deprecated. Not supported yet.
round(double, DoubleUtils.RoundingScheme) - Static method in class com.numericalmethod.suanshu.number.DoubleUtils
Round up or down a double to an integer.
round(double, int) - Static method in class com.numericalmethod.suanshu.number.DoubleUtils
Round a double to the precision specified.
rowEchelonForm(Matrix, double) - Static method in class com.numericalmethod.suanshu.matrix.doubles.IsMatrix
Check if a matrix is in the row echelon form.
rows(Matrix, int[]) - Static method in class com.numericalmethod.suanshu.matrix.doubles.operation.CreateMatrix
CreateMatrix a sub-matrix from the rows of a matrix.
rows(Matrix, int, int) - Static method in class com.numericalmethod.suanshu.matrix.doubles.operation.CreateMatrix
CreateMatrix a sub-matrix from the rows of a matrix.
rowSums(Matrix) - Static method in class com.numericalmethod.suanshu.matrix.doubles.operation.MatrixUtils
 
RSS - Variable in class com.numericalmethod.suanshu.stats.regression.linear.ols.Residuals
diagnostic measure: the sum of squared residuals, Σ(ε^2)
run(T) - Method in interface com.numericalmethod.suanshu.parallel.IterationBody
 
run(int) - Method in interface com.numericalmethod.suanshu.parallel.LoopBody
This method contains the code inside the for-loop, as in a native for-loop like this: for (int i = start; i < end; i += increment) { // loop body }
run(int, AutoCovarianceFunction) - Method in class com.numericalmethod.suanshu.stats.timeseries.linear.multivariate.stationaryprocess.InnovationAlgorithmImpl
Run the Innovation Algorithm to compute the prediction parameters.

SuanShu, a Java numerical and statistical library
A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
Copyright © 2011 Numerical Method Inc. Ltd. All Rights Reserved.