Changes in version 1.2.1 (2026-03-28)                  

  - Fixed an issue related to floating point arithmetic (#51)

                 Changes in version 1.2.0 (2025-08-22)                  

  - Implemented efficient algorithms for roll_median and roll_quantile
    functions

  - Fixed an issue for weights argument if single value (#45)

                 Changes in version 1.1.7 (2024-04-05)                  

  - New roll_crossprod function for computing rolling and expanding
    crossproducts of time-series data

  - Restructured the C++ header files to be used by other packages

                 Changes in version 1.1.5 (2020-07-05)                  

  - Added hex sticker and improved documentation (#27, #28, #30)

  - New roll_quantile function for computing rolling and expanding
    quantiles of time-series data
    
      - Note: roll_quantile function is not calculated using an online
        algorithm

  - Fixed an issue in the roll_min and roll_max functions (#32)

                 Changes in version 1.1.4 (2020-01-17)                  

  - Fixed issues related to floating point arithmetic (#23, #24)

                 Changes in version 1.1.3 (2019-10-04)                  

  - Added vector support to each function (#20)

  - Implemented efficient algorithms for roll_min and roll_max functions

  - New roll_idxmin and roll_idxmax functions for computing rolling and
    expanding indices of minimums and maximums, respectively, of
    time-series data (#22)

                 Changes in version 1.1.2 (2019-02-06)                  

  - New roll_median, roll_min, roll_max, roll_any, and roll_all
    functions for computing rolling and expanding medians, minimums,
    maximums, any, and all, respectively, of time-series data (#4, #13,
    #14)
    
      - Note: roll_median, roll_min, and roll_max functions are not
        calculated using online algorithms

                 Changes in version 1.1.1 (2018-08-11)                  

  - Added online argument to process observations using online
    algorithms by default

  - roll_lm function now returns standard errors (#7)

  - Simplified checks for width and min_obs arguments (#3)

  - Added y argument to roll_cov and roll_cor functions (#2)

  - Deprecated less common functions (roll_eigen, roll_vif, and
    roll_pcr) and arguments (scale and center in the roll_lm function);
    also removed the parallel_for argument in favor of a new approach
    used internally

                 Changes in version 1.0.7 (2017-05-01)                  

  - New roll_sum and roll_prod functions for computing rolling and
    expanding sums and products, respectively, of time-series data

                 Changes in version 1.0.6 (2016-09-19)                  

  - Added intercept argument to roll_lm and roll_pcr functions

                 Changes in version 1.0.5 (2016-08-29)                  

  - Fixed an issue in the src/Makevars and src/Makevars.win files (#1)

                 Changes in version 1.0.4 (2016-07-05)                  

  - roll_lm and roll_pcr functions have been enhanced:
    
      - y can now be a matrix or xts object with multiple dependent
        variables
    
      - Added shorthand arguments for center and scale

  - New roll_scale function for computing rolling and expanding scaling
    and centering of time-series data