Data Engineering and BI courses are free this week!
Re-arrange the multi-step residuals
arrange_hres(list_res)A vector or a matrix of multi-step residuals
a list of H multi-step residuals. Each element of the list can be a vector (univariate time series) or a matrix (multivariate time series).
Let Z_t, t=1,...,T, be a univariate time series. We can define the multi-step residuals such us _h,t = Z_t+h - Z_t+h|t h t T-h where Z_t+h|t is the h-step fitted value, calculated as the h-step ahead forecast given the time t. Given the list of errors at different step ([_1,1, \; ..., \; _1,T], ..., [_H,1, \; ..., \; _H,T]) this function returns a T-vector with the residuals, organized in the following way: [_1,1 \; _2,2 \; ... \; _H,H \; _1,H+1 \; ... \; _H,T-H]' Same idea can be apply for a multivariate time series.
Other utilities:
Cmatrix(),
FoReco2ts(),
agg_ts(),
commat(),
ctf_tools(),
hts_tools(),
lcmat(),
oct_bounds(),
residuals_matrix(),
score_index(),
shrink_estim(),
thf_tools()