TSstudio

Installation

install.packages("TSstudio")

# install.packages("devtools")devtools::install_github("RamiKrispin/TSstudio")

Usage

Plotting time series data

library(TSstudio)data(USgas)# Ploting time series objectts_plot(USgas,title="US Monthly Natural Gas Consumption",Ytitle="Billion Cubic Feet")

Seasonality analysis

# Seasonal plotts_seasonal(USgas,type="all")

# Heatmap plotts_heatmap(USgas)

Correlation analysis

# ACF and PACF plotsts_cor(USgas,lag.max=60)

# Lags plotts_lags(USgas,lags=1:12)

# Seasonal lags plotts_lags(USgas,lags= c(12,24,36,48))

Training forecasting models

# Forecasting applications# Setting training and testing partitionsUSgas_s<- ts_split(ts.obj=USgas,sample.out=12)train<-USgas_s$traintest<-USgas_s$test# Forecasting with auto.arimalibrary(forecast)md<- auto.arima(train)fc<- forecast(md,h=12)# Plotting actual vs. fitted and forecastedtest_forecast(actual=USgas,forecast.obj=fc,test=test)

# Plotting the forecastplot_forecast(fc)

# Run horse race between multiple modelsmethods<-list(ets1=list(method="ets",method_arg=list(opt.crit="lik"),notes="ETS model with opt.crit = lik"),ets2=list(method="ets",method_arg=list(opt.crit="amse"),notes="ETS model with opt.crit = amse"),arima1=list(method="arima",method_arg=list(order= c(2,1,0)),notes="ARIMA(2,1,0)"),arima2=list(method="arima",method_arg=list(order= c(2,1,2),seasonal=list(order= c(1,1,1))),notes="SARIMA(2,1,2)(1,1,1)"),hw=list(method="HoltWinters",method_arg=NULL,notes="HoltWinters Model"),tslm=list(method="tslm",method_arg=list(formula=input~trend+season),notes="tslm model with trend and seasonal components"))# Training the models with backtestingmd<- train_model(input=USgas,methods=methods,train_method=list(partitions=6,sample.out=12,space=3),horizon=12,error="MAPE")# A tibble: 6 x 7model_idmodelnotesavg_mapeavg_rmse`avg_coverage_80%``avg_coverage_95%`<chr><chr><chr><dbl><dbl><dbl><dbl>1arima2arima       SARIMA(2,1,2)(1,1,1)0.0557167.0.5830.8062hwHoltWintersHoltWintersModel0.0563163.0.7360.8893ets1etsETSmodelwithopt.crit=lik0.0611172.0.6810.9034ets2etsETSmodelwithopt.crit=amse0.0666186.0.4580.8335tslmtslmtslmmodelwithtrendandseasonalcomponents0.0767220.0.4170.6676arima1arima       ARIMA(2,1,0)0.188598.0.8750.958

# Plot the performance of the different models on the testing partitionsplot_model(md)

# Holt-Winters tunning parameters with grid searchhw_grid<- ts_grid(USgas,model="HoltWinters",periods=6,window_space=6,window_test=12,hyper_params=list(alpha= seq(0,1,0.1),beta= seq(0,1,0.1),gamma= seq(0,1,0.1)))                                       plot_grid(hw_grid,type="3D")