- Fixed an issue related to floating point arithmetic (#51)
-
Implemented efficient algorithms for
roll_medianandroll_quantilefunctions -
Fixed an issue for
weightsargument if single value (#45)
-
New
roll_crossprodfunction for computing rolling and expanding crossproducts of time-series data -
Restructured the C++ header files to be used by other packages
-
Added hex sticker and improved documentation (#27, #28, #30)
-
New
roll_quantilefunction for computing rolling and expanding quantiles of time-series data- Note:
roll_quantilefunction is not calculated using an online algorithm
- Note:
-
Fixed an issue in the
roll_minandroll_maxfunctions (#32)
-
Added vector support to each function (#20)
-
Implemented efficient algorithms for
roll_minandroll_maxfunctions -
New
roll_idxminandroll_idxmaxfunctions for computing rolling and expanding indices of minimums and maximums, respectively, of time-series data (#22)
-
New
roll_median,roll_min,roll_max,roll_any, androll_allfunctions for computing rolling and expanding medians, minimums, maximums, any, and all, respectively, of time-series data (#4, #13, #14)- Note:
roll_median,roll_min, androll_maxfunctions are not calculated using online algorithms
- Note:
-
Added
onlineargument to process observations using online algorithms by default -
roll_lmfunction now returns standard errors (#7) -
Simplified checks for
widthandmin_obsarguments (#3) -
Added
yargument toroll_covandroll_corfunctions (#2) -
Deprecated less common functions (
roll_eigen,roll_vif, androll_pcr) and arguments (scaleandcenterin theroll_lmfunction); also removed theparallel_forargument in favor of a new approach used internally
- New
roll_sumandroll_prodfunctions for computing rolling and expanding sums and products, respectively, of time-series data
- Added
interceptargument toroll_lmandroll_pcrfunctions
- Fixed an issue in the
src/Makevarsandsrc/Makevars.winfiles (#1)
-
roll_lmandroll_pcrfunctions have been enhanced:-
ycan now be a matrix or xts object with multiple dependent variables -
Added shorthand arguments for
centerandscale
-
-
New
roll_scalefunction for computing rolling and expanding scaling and centering of time-series data