<< aggreg1 Time series disaggregation bfl1 >>

Grocer >> Time series disaggregation > bfl

bfl

Temporal disaggregation using the Boot-Feibes-Lisman method

CALLING SEQUENCE

[y,res]=bfl(nameY,ta,d,s)

PARAMETERS

Input

• nameY = a ts or a Nx1 vector of low frequency data or a string representing the name of a vector or a ts

• ta = type of disaggregation

 - ta=-1 ---> sum (flow)

 - ta=0 ---> average (index)

 - ta=i ---> i th element (stock) ---> interpolation

• d = objective function to be minimized: volatility of - - d=0 ---> levels

 - d=1 ---> first differences

 - d=2 ---> second differences

• s = number of high frequency data points for each low frequency data point

 - s= 4 ---> annual to quarterly

 - s=12 ---> annual to monthly

 - s= 3 ---> quarterly to monthly

Output

• y = High frequency estimate

• res = a results tlist with:

  - res('meth') = 'Boot-Feibes-Lisman'

  - res('nobs_lf') = Number of low frequency data

  - res('aggreg_mode') = Type of disaggregation

  - res('s') = Frequency conversion

  - res('diff') = Degree of differencing

  - res('y_lf') = Low frequency data

  - res('y') = High frequency estimate

  - res('namey') = name of the low frequency data

DESCRIPTION

Provides temporal disaggregation of low frequency data using the Boot-Feibes-Lisman method.

EXAMPLE

load(GROCERDIR+'\data\xesp.dat')
[y,res] = bfl(Y,-1,1,12);
 
// Example taken from function bfl_d. Provides monthly disaggregation of series Y (in bfl_d a vector, but it should also work if Y were a ts) using first differences and with the annual data equal of the sum of corresponding monthly data.

AUTHOR

Eric Dubois 2005

<< aggreg1 Time series disaggregation bfl1 >>