<< deming Single equation regressions dynfore >>

Grocer >> Single equation regressions > deming1

deming1

deming linear estimation

CALLING SEQUENCE

res = deming1(y,x,lambda,alpha)

PARAMETERS

Input

* y = (N x 1) vector of the endogenous variable measured with error

* x = (N x 1) vector of the endogenous variable measured with error

* lambda = a scalar, the assumed ratio of the measurement errors of y and x: sigma2_y / sigma2_x

* alpha = a scalar, the confidence level

Output

* res = a results tlist with

   - res('meth') = 'deming'

   - res('y') = the y vector of endogenous variables

   - res('x') = the x vector of exogenous variables

   - res('errors variance ratio') = the assumed ratio of the measurement errors of y and x

   - res('Estimated x values') = the estimated "true" x values

   - res('Estimated y values') = the estimated "true" y values

   - res('regression st. error') = Standard error of regression estimate

   - res('coeff st. error') = estimate of standard error of the slope and intercept

   - res('tstat') = Sutnedt's T of the estimated coefficients

   - res(string(100*alpha)+'% confidence interval') = the confidence interval of the coefficients at the chosen level

DESCRIPTION

performs a linear Deming regression to find the linear coefficients: y = b(1) + b(2)*x under the assumptions that x and y *both* contain measurement error with measurement error variance related as lambda = sigma2_y/sigma2_x (sigma2_x and sigma2_y is the measurement error variance of the x and y variables, respectively)..

EXAMPLE

// provides the deming regression presented at:
// http://www.real-statistics.com/regression/deming-regression/deming-regression-basic-concepts/
 
load(GROCERDIR+'\deming_d.dat')
res = deming1(y,x,0.4,0.95)

AUTHOR

Éric Dubois 2018

Report an issue
<< deming Single equation regressions dynfore >>