Factor density estimation using bootstrapping
res = fac_density_boot(arg1,...,argn)
* arg1,...,argn = arguments which can be:
. a time series
. a real (n x p) vector
. a string equal to the name of a time series or of a (n x p) real vector 'stan' if the user wants to perform the analysis on standardized variables
. 'nb_samples=xx' where xx is the number of samples drawn
. 'noprint' wif the user does not want to plor the results
* res = a results tlist with
- res('meth') = 'factor density via bootstrapping'
- res('y') = the (nobs x k) matrix collecting the variables
- res('standardization') = a boolean, indicating whether variables have been standardized
- res('# of draws') = an integrer, the # of draws
- res('factors') = a (nobs x n_samples) matrix, the factor density
- res('factors percentiles') = a (nobs x 5) matrix, the percentiles (2.5, 25, 50, 75 and 97.5)
- res('namey') = name of the variables
- res('prests') = a boolean indicating the presence or absence of a time series in the variables list
- res('bounds') = if there is a timeseries in the regression, the bounds of the regression
global GROCERDIR load(GROCERDIR+'\data\esp_yields.dat') // provides the result of the resampling by the bootstrapping method on factors estimation on Spanish yields data, standardized, with 1000 draws: res = fac_density_lko(esp_yields,'stan'); // provides the result of the resampling by the leave k-out methods on factors estimation on Spanish yields data, standardized, with 10000 draws: res = fac_density_lko(esp_yields,'stan','n_draws=10000'); |  |  |