Nonparametric Estimation of the Trend and Its Derivatives in TS
Forecasting Function for ARMA Models via Bootstrap
Asymptotically Unbiased Confidence Bounds
ARMA Order Selection Matrix
Data-driven Local Polynomial for the Trend's Derivatives in Equidistan...
Extract Model Fitted Values
Estimation of Trends and their Derivatives via Local Polynomial Regres...
Estimation of Nonparametric Trend Functions via Kernel Regression
Forecasting Function for Trend-Stationary Time Series
Data-driven Nonparametric Regression for the Trend in Equidistant Time...
Forecasting Function for ARMA Models under Normally Distributed Innova...
Optimal Order Selection
Plot Method for the Package 'smoots'
Print Method for the Package 'smoots'
Rescaling Derivative Estimates
Extract Model Residuals
Backtesting Semi-ARMA Models with Rolling Forecasts
smoots: A package for data-driven nonparametric estimation of the tren...
Forecasting Function for Nonparametric Trend Functions
Advanced Data-driven Nonparametric Regression for the Trend in Equidis...
The nonparametric trend and its derivatives in equidistant time series (TS) with short-memory stationary errors can be estimated. The estimation is conducted via local polynomial regression using an automatically selected bandwidth obtained by a built-in iterative plug-in algorithm or a bandwidth fixed by the user. A Nadaraya-Watson kernel smoother is also built-in as a comparison. With version 1.1.0, a linearity test for the trend function, forecasting methods and backtesting approaches are implemented as well. The smoothing methods of the package are described in Feng, Y., Gries, T., and Fritz, M. (2020) <doi:10.1080/10485252.2020.1759598>.