aic5()
|
aic5: slightly faster model estimations
todo: rewrite backcast and aic.wge for real speed gains
returns the top 5 aic |
aicbic()
|
aicbic function
returns the top 5 aic and bic values for an ARMA time series |
ase()
|
ASE calculator (specific to tswge)
Takes the last length(xhat) entries in x and compares them to xhat |
assess()
|
assessment function!
assess a time series |
calculate_ar1_varx()
|
Computes the Population Variance of an AR(1) Time Series |
calculate_arp_varx()
|
Computes the Population Variance of an AR(p) Time Series
Note that this can also be used for AR(1) models instead of using calculate_ar1_varx
although, it will need an extra argument 'pt' |
calculate_ts_gamma()
|
Computes Gamma (Auto Covariance of a time series realization) |
calculate_ts_gamma0()
|
Computes Gamma0 (Variance of a time series realization) |
calculate_ts_mean()
|
Computes the Mean of a Time Series realization |
calculate_ts_mean_confidence_interval()
|
Computes the Confidence Interval of the Mean of a Time Series |
calculate_ts_rho()
|
Computes Rho (Auto Correlation of a time series realization) |
calculate_ts_var_of_mean()
|
Computes the Variance of the Sampling distribution of the Mean of a Time Series |
check_stationarity()
|
Wrapper to check the stationarity of a Time Series |
compute_a()
|
Given an AR(p,q) realization, this function estimates the white
noise estimates using equation 6.23 (from the text book) |
compute_stda()
|
Given the white noise estimates, this function computes
the standard deviation of the white noise. Note that non-zero
terms are removed before calcualting the variance. |
compute_vara()
|
Given the white noise estimates, this function computes
the variance of the white noise. Note that non-zero terms
are removed before calcualting the variance. |
difference()
|
Time Series transformation |
estimate()
|
paramater estimation function |
evaluate_residuals()
|
Evaluate if the dats is consistent with white noise |
factor.wge.season()
|
Prints the factor table for a pure seasonal model (p = d = q = 0) |
fcst()
|
forecast function! |
generate()
|
generator function! |
generate_multiple_realization()
|
Generate Multiple Realizations of a model
Useful for checking model appropriateness |
get_all_a_calc()
|
Given an AR(p,q) realization, this function estimates the white
noise estimates using equation 6.23 (from the text book), the
variance and the standard deviation of the white noise |
hush()
|
Prevents inter 'cat' statememt from printing |
lag_dfr()
|
lag_dfr: Lag columns of a data frame |
ljung_box()
|
ljung box test for white noise |
mlag_dfr()
|
mlag_dfr: Lag multiple columns of a data frame |
ModelBuildMultivariateVAR
|
R6 class ModelBuildMultivariateVAR |
ModelBuildNNforCaret
|
R6 class ModelBuildNNforCaret |
ModelCombine
|
R6 class ModelCompareMultivariate |
ModelCompareBase
|
R6 class ModelCompareBase |
ModelCompareMultivariateVAR
|
R6 class ModelCompareMultivariate |
ModelCompareNNforCaret
|
R6 class ModelCompareMultivariate |
ModelCompareUnivariate
|
R6 class ModelCompareUnivariate |
model_cor()
|
Model Correlated data
Interactively build a model of correlated data |
model_det()
|
model with deterministic signal plus noise |
model_mlr()
|
Model as MLR model |
MultivariateEDA
|
R6 class MultivariateEDA |
overfit()
|
overfit function
produces an overfit table |
phicheck()
|
check if ar component is stationary (stolen from the R stats source code) |
playground()
|
Playground: generate a random time series for practice |
plot_multiple_realizations()
|
Plot multiple realizations of a model
Useful for checking model appropriateness |
plot_res()
|
plot the residuals of a time series |
rarima()
|
radnom arima |
rarma()
|
random arma |
raruma()
|
random aruma |
rsigpn()
|
random signal plus noise |
sliding_ase_univariate()
|
Function to calculate the sliding window ASE for a model
Supports ARMA, ARIMA, ARUMA (seasonal ARIMA) and Signal Plus Noise Models |
sliding_ase_var()
|
Function to calculate the sliding window ASE for a model
Supports VAR Model from the vars package |