Evaluate if the dats is consistent with white noise

evaluate_residuals(
  x,
  p = 0,
  q = 0,
  k_val = c(24, 48),
  alpha = 0.05,
  lag.max = 50,
  model_name = "My Model"
)

Arguments

x	the time series
p	the ar order (Default = 0)
q	the ma order (Default = 0)
k_val	a vector of k values for ljung_box test
alpha	Significance level to be used for ljung_box tests
lag.max	Value of lags to plot for the ACF (Default: 50)
model_name	Model name or identifier (Default: "My Model")

Value

the results of the tests, in tidy data format

Examples

library(tswge)
#> Warning: package 'tswge' was built under R version 3.5.3

# Generated White Noise 
wn = gen.arma.wge(n = 200, sn = 101)
evaluate_residuals(wn)
#> Loading required namespace: ggfortify
#> Loading required namespace: patchwork
#> None of the 'ljung_box' tests rejected the null hypothesis that the data is consistent with white noise at an significance level of  0.05  
#> # A tibble: 2 x 7
#>   test               K chi.square    df  pval Model    Decision
#>   <chr>          <dbl>      <dbl> <dbl> <dbl> <chr>    <chr>   
#> 1 Ljung-Box test    24       19.6    24 0.719 My Model FTR NULL
#> 2 Ljung-Box test    48       41.2    48 0.747 My Model FTR NULL

# Not White Noise
data(hadley)
evaluate_residuals(hadley)
#> At least one of the 'ljung_box' tests rejected the null hypothesis that the data is consistent with white noise at an significance level of  0.05  
#> # A tibble: 2 x 7
#>   test               K chi.square    df  pval Model    Decision   
#>   <chr>          <dbl>      <dbl> <dbl> <dbl> <chr>    <chr>      
#> 1 Ljung-Box test    24      1315.    24     0 My Model REJECT NULL
#> 2 Ljung-Box test    48      1469.    48     0 My Model REJECT NULL