Transform Univariate Time Series
Score transformation
Selecting lambda
Skewness/Kurtosis Value
Standarization
Removes measure of centrality from the series
Compute lagged differnces
Deterministic Trend
Fill with "linear approximation"
Fill with "Last Observation Carried Forward"
Fill with "Next observation carried backwards"
Fill with "cubic spline interpolation"
Baxter-King Filter
Butterworth Filter
Christiano-Fitzgerald Filter
Hamilton Filter
Hodrick-Prescot Filter
Trigonometric regression Filter
Geometric Mean value
Compute lagged or leading values
Mode value
Detect outliers with Tukey's method
Detect outliers with Percentiles
Detect outliers with zscore
Detect outliers Iglewicz and Hoaglin (1993) robust z-score method
Detect outliers with upper and lower threshold
Winsorize
Plotting wrapper around plot.default
nth Power Transformation
Box-Cox Transformations
Manly(1971) Transformations
Tukey Transformations Transformations
Yeo and Johnson(2000) Transformations
Change the base year
nth Root Transformation
Rescale
Univariate time series operations that follow an opinionated design. The main principle of 'transx' is to keep the number of observations the same. Operations that reduce this number have to fill the observations gap.