The R Project SVN R

Rev

Rev 27497 | Rev 30412 | Go to most recent revision | Show entire file | Ignore whitespace | Details | Blame | Last modification | View Log | RSS feed

Rev 27497 Rev 28101
Line 38... Line 38...
38 
 evolves by
 38 
 evolves by
39 
 \deqn{\mu_{t+1} = \mu_t + \xi_t,  \qquad \xi_t \sim N(0, \sigma^2_\xi)}{%
 39 
 \deqn{\mu_{t+1} = \mu_t + \xi_t,  \qquad \xi_t \sim N(0, \sigma^2_\xi)}{%
40 
   m[t+1] = m[t] + xi[t], xi[t] ~ N(0, sigma^2_xi)}
 40 
   m[t+1] = m[t] + xi[t], xi[t] ~ N(0, sigma^2_xi)}
41 
 The observations are
 41 
 The observations are
42 
 \deqn{x_t = \mu_t + \epsilon_t, \qquad \epsilon_t \sim  N(0, \sigma^2_\epsilon)}{%
 42 
 \deqn{x_t = \mu_t + \epsilon_t, \qquad \epsilon_t \sim  N(0, \sigma^2_\epsilon)}{%
43 
   x[t] = m[t] + eps[t], exp[t] ~  N(0, sigma^2_eps)}
 43 
   x[t] = m[t] + eps[t], eps[t] ~  N(0, sigma^2_eps)}
44 
 There are two parameters, \eqn{\sigma^2_\xi}{sigma^2_xi}
 44 
 There are two parameters, \eqn{\sigma^2_\xi}{sigma^2_xi}
45 
 and \eqn{\sigma^2_\epsilon}{sigma^2_eps}.  It is an ARIMA(0,1,1) model,
 45 
 and \eqn{\sigma^2_\epsilon}{sigma^2_eps}.  It is an ARIMA(0,1,1) model,
46 
 but with restrictions on the parameter set.
 46 
 but with restrictions on the parameter set.
47 
 
 47 
 
48 
 The \emph{local linear trend model}, \code{type = "trend"}, has the same
 48 
 The \emph{local linear trend model}, \code{type = "trend"}, has the same