<< denton1 Time series disaggregation difonzo1 >>

Grocer >> Time series disaggregation > difonzo

difonzo

Multivariate temporal disaggregation with transversal constraint

CALLING SEQUENCE

[y,res]=difonzo(namey,namex,namez,d,arg1,...,argn)

PARAMETERS

Input

* nameY = a (N x M) matrix, a list of M vectors or ts or a string vector representing names such objects

    ---> M series of low frequency data with N observations

* namex = a (n x M) matrix, a list of M vectors or ts or a string vector representing names such objects

     ---> M series of high frequency data with n observations

* namez = a (n x 1) vector or a ts

     ---> high frequency transversal constraint

* d = objective function to be minimized: volatility of

  - d=0 ---> levels

  - d=1 ---> first differences

  - d=2 ---> second differences

* argi = an argument which can be

  - the string 'divfq=n' where n is the number of high frequency data points for each low frequency data points

  - the string 'ta=n' where n is the aggregation type:

      . n=-1 (default) ---> sum (flow)

      . n=0 ---> average (index)

      . n=i ---> i th element (stock) ---> interpolation

  - the string 'typemod=n' where n is either 'wn' for the multivariare white noise model (default) or 'rw' for the multivariate random walk model

 

Output

* y = High frequency estimate

* res = a results tlist with:

  - res('meth') = 'Multivariate difonzo'

  - res('typemod') = type of the model for the high frequency innovations

  - res('ta') = type of disaggregation

  - res('nobs_lf') = nobs. of low frequency data

  - res('nobs_hf') = nobs. of high-frequency data

  - res('pred') = number of extrapolations

  - res('s') = frequency conversion between low and high freq.

  - res('diff') = Degree of differencing

  - res('y') = high frequency estimate

  - res('y_lf') = low frequency data

  - res('indicator') = high frequency indicators

  - res('transversal') = data for the transversal constraint

  - res('y_dt') = high frequency estimate: standard deviation

  - res('resid') = high frequency residuals

  - res('resid_U') = low frequency residuals

  - res('beta') = estimated model parameters

  - res('sd') = standard deviation of the estimated model parameters

  - res('namey') = Name of the high frequency aggregate

  - res('namex') = Name of the low frequency indicators

  - res('namez') = Name of the high frequency transversal constraint

DESCRIPTION

Temporal disaggregation using the difonzo method of multivariate temporal disaggregation with transversal constraint (high level function with vectors, matrices or ts and the possibility of default parameters).

EXAMPLE

[y,res] = difonzo(Y,x,z,'ta=0','typemod=rw')
 // Example taken from function difonzo_d. Provides quarterly disaggregation of (23 x 2) matrix Y (representing 2 annual data)
// with (92x2) matrix x (representing 2 quarterly series) and transversality constraint represented by the (92x1) matrix z.
// Annual series are built by averaging quarterly series (ta=0); the order of aggregation is determined by the function;
// residuals are assumed to follow a multivariate random walk ('typemod=rw').

AUTHOR

Eric Dubois 2005

Report an issue
<< denton1 Time series disaggregation difonzo1 >>