tsutils (version 0.9.0)

trendtest: Test a time series for trend

Description

Test a time series for trend by either fitting exponential smoothing models and comparing then using the AICc, or by using the non-parametric Cox-Stuart test. The tests can be augmented by using multiple temporal aggregation.

Usage

trendtest(y, extract = c("FALSE", "TRUE"), type = c("aicc", "cs"),
  mta = c(FALSE, TRUE))

Arguments

y

a time series that must be of either ts or msts class.

extract

if TRUE then the centred moving average of the time series is calculated and the test is performed on that. Otherwise, the test is performed on the raw data.

type

type of test. Can be:

• "aicc": test by comparing the AICc of exponential smoothing models. See details.

• "cs": test by using the Cox-Stuart test. See details.

mta

If TRUE augment testing by using Multiple Temporal Aggregation.

Value

The function returns TRUE when there is evidence of trend and FALSE otherwise.

Details

All tests are performed at 5

References

The multiple temporal aggregation follows the construction approach suggested by Kourentzes, N., Petropoulos, F., & Trapero, J. R. (2014). Improving forecasting by estimating time series structural components across multiple frequencies. International Journal of Forecasting, 30(2), 291-302.

Examples

# NOT RUN {
trendtest(referrals,TRUE)

# }