tests/testthat/test-arima.R

In pli2016/forecast: Forecasting Functions for Time Series and Linear Models

# A unit test for Arima() function
if(require(testthat))
{
	context("Tests on input")
	test_that("tests for a non-ts object", {
  	set.seed(123)
  	abc <- rnorm(50, 5, 1)
  	fit <- Arima(abc, order = c(2, 0, 1))
  	expect_that(fit$arma, equals(c(2, 1, 0, 0, 1, 0, 0)))
	})

	test_that("tests for a ts with the seasonal component", {
  	fit <- Arima(wineind, order = c(1, 1, 1), seasonal = c(0, 1, 1))
  	expect_that(fit$arma, equals(c(1, 1, 0, 1, 12, 1, 1)))
	})

	test_that("tests for ARIMA errors", {
  	fit <- Arima(wineind, order = c(1, 1, 1), seasonal = c(0, 1, 1))
  	expect_that(arima.errors(fit), equals(wineind))
	})

	test_that("tests for arimaorder", {
  	for(ar in 1:5){
  		for(i in 0:1){
    			for(ma in 1:5){
			      fitarima <- Arima(lynx, order = c(ar, i, ma), method = "ML", include.constant = TRUE, lambda = 0.5)
			      arextracted <- fitarima$arma[1]
			      iextracted <- fitarima$arma[6]
			      maextracted <- fitarima$arma[2]
			      expect_true(all(arimaorder(fitarima) == c(arextracted, iextracted, maextracted)))
		    }
		  }
	}
	})

	test_that("tests for forecast.Arima", {
	fit1 <- Arima(wineind, order = c(1, 1, 2), seasonal = c(0, 1, 1), method = "CSS")
	expect_error(forecast.Arima(fit1, xreg = 1:10))
	expect_warning(forecast.Arima(fit1, include.drift = TRUE))
	expect_true(all.equal(forecast.Arima(fit1, bootstrap = TRUE, npaths = 100)$ mean, forecast.Arima(fit1)$mean))

	fit2 <- Arima(wineind, order = c(1, 0, 1), seasonal = c(0, 0, 0), include.drift = TRUE)
	expect_warning(Arima(wineind, order = c(1, 2, 1), include.drift = TRUE))
	expect_true("drift" %in% names(coef(fit2)))
	expect_true(length(forecast.Arima(fit2)$mean) == 2 * frequency(wineind))

	fit3 <- Arima(wineind, order = c(1, 1, 2), seasonal = c(0, 1, 1), include.mean = FALSE)
	expect_false("intercept" %in% names(coef(fit3)))
	expect_true(frequency(forecast.Arima(fit3)$mean) == frequency(wineind))

	fit4 <- Arima(wineind, order = c(1, 1, 2), seasonal = c(0, 1, 1), xreg = rnorm(length(wineind)))
	expect_error(forecast.Arima(fit4))
	expect_error(forecast.Arima(fit4, xreg = matrix(rnorm(40), ncol = 2)))
	expect_true(length(forecast.Arima(fit4, xreg = rnorm(20))$mean) == 20)

	fit5 <- Arima(wineind[1:150], order = c(1, 1, 2), seasonal = c(0, 1, 1), method = "ML")
	expect_true(accuracy(fit5)[1, "MAPE"] < accuracy(Arima(wineind, model = fit5))[1, "MAPE"])

	fit6 <- Arima(wineind, order = c(1, 1, 2), seasonal = c(0, 1, 1), method = "CSS", lambda=5)
	expect_false(identical(fit1$coef, fit6$coef))
	})

	test_that("tests for search.arima", {
	set.seed(444)
	arimasim <- arima.sim(n = 300, model = list(ar = runif(8, -.1, 0.1), ma = runif(8, -0.1, 0.1), sd = 0.1))
	expect_true(AIC(auto.arima(arimasim)) >= AIC(auto.arima(arimasim, stepwise = FALSE)))
	})

	test_that("tests for forecast.ar()", {
	fitar <- ar(taylor)
	arfc <- forecast.ar(fitar)$mean
	expect_true(all(arfc == forecast.ar(fitar, bootstrap = TRUE, npaths = 100)$mean))
	expect_true(all(arfc == forecast.ar(fitar, fan = TRUE)$mean))
	expect_error(forecast.ar(fitar, level = -10))
	expect_error(forecast.ar(fitar, level = 110))
	expect_true(all(arfc + 1 == forecast.ar(fitar, lambda = 1)$mean))
	arfcbc <- forecast.ar(fitar, lambda=2)
	arfcabc <- forecast.ar(fitar, lambda=2, biasadj = TRUE)
	expect_false(identical(arfcbc$mean, arfcabc$mean))
	})

	test_that("tests for as.character.Arima()", {
	expect_match(as.character(auto.arima(woolyrnq)), regexp = "ARIMA")
	})
}