APPENDIX A ABBREVIATIONS rBe Recursive Bayesian estimation OLS ordinary least squares WLS Weighted least squares DLS Discounted least squares GLS Generalised least squares 2SLS Two stage least squares 3SLS Three stage least squares FIML Full information maximum likelihood kt Kilotonnes ; 1 kilotonne = 1 million kilograms DSSE Dynamic sum of squared errors RMDSSE Root mean dynamic sum of squared errors RMSE Root mean squared error SEE Standard error of estimate KF Kalman filter 'When I use a word,' Humpty Dumpty said in a rather scornful tone, 'it means just what I choose it to mean - neither more nor less' Through the Looking Glass, Lewis Carroll APPENDIX B NOTATION The following notation is used throughout the paper. p(A) The probability of A p(A AND B) The probability of A and B p(A GIVEN B) The conditional probability of A, given B L(A GIVEN B) The likelihood of A, given B E(X) The expectation (mean) of X Var(X) The variance of X. If X is a vector, the variance-covariance matrix of X (also called the variance matrix, the covariance matrix or the dispersion matrix). Prec(X) The precision of X, =(Var(X))-1, the inverse of the variance matrix. Also called the information matrix. SUMnt(Xt) The sum of Xt over t = 1 to n. SUMdt(X) The integral of X with respect to t. XE An estimate of X XF A forecast of X XO An observation of X A is N(M,C) A is distributed normally with mean M and variance-covariance matrix C. A is Beta(m,n) A is distributed as a Beta distribution with parameters m, n. I The identity matrix AT The transpose of the matrix A A-1 The inverse of the matrix A AE-1 The inverse of the estimate of A Y The dependent variable (or vector of variables) X The independent variable (or vector or matrix of variables) B The model parameter (or vector of parameters) Yt The dependent variable at time t Datat(Y) The entire sequence of variables Y1, Y2, ...Yt G The gain of a filter H The systematic component of parameter changes V The variance of the stochastic component of the dependent variable. W The variance of the stochastic component of the parameter changes. v Random Normal vectors distributed N(0,V) w Random Normal vectors distributed N(0,W) NDC Net domestic consumption of wool PW Price of wool PS Price of synthetic fibre POP Population PCE Personal consumption expenditure