################################################################################
# Copyright 2017-2018 Gabriele Valentini, Douglas G. Moore. All rights reserved.
# Use of this source code is governed by a MIT license that can be found in the
# LICENSE file.
################################################################################
library(rinform)
context("Predictive Information")
test_that("predictive_info checks parameters", {
xs <- sample(0:1, 10, T)
expect_error(predictive_info("series", kpast = 1, kfuture = 1, local = !T))
expect_error(predictive_info(NULL, kpast = 1, kfuture = 1, local = !T))
expect_error(predictive_info(NA, kpast = 1, kfuture = 1, local = !T))
expect_error(predictive_info(xs, kpast = "k", kfuture = 1, local = !T))
expect_error(predictive_info(xs, kpast = NULL, kfuture = 1, local = !T))
expect_error(predictive_info(xs, kpast = NA, kfuture = 1, local = !T))
expect_error(predictive_info(xs, kpast = 0, kfuture = 1, local = !T))
expect_error(predictive_info(xs, kpast = -1, kfuture = 1, local = !T))
expect_error(predictive_info(xs, kpast = 1, kfuture = "k", local = !T))
expect_error(predictive_info(xs, kpast = 1, kfuture = NULL, local = !T))
expect_error(predictive_info(xs, kpast = 1, kfuture = NA, local = !T))
expect_error(predictive_info(xs, kpast = 1, kfuture = 0, local = !T))
expect_error(predictive_info(xs, kpast = 1, kfuture = -1, local = !T))
expect_error(predictive_info(xs, kpast = 1, kfuture = 1, local = "TRUE"))
expect_error(predictive_info(xs, kpast = 1, kfuture = 1, local = NULL))
expect_error(predictive_info(xs, kpast = 1, kfuture = 1, local = NA))
})
test_that("predictive_info on single series", {
expect_equal(predictive_info(c(1, 1, 0, 0, 1, 0, 0, 1),
kpast = 2, kfuture = 1, local = !T),
0.918296, tolerance = 1e-6)
expect_equal(predictive_info(c(0, 0, 1, 1, 1, 1, 0, 0, 0),
kpast = 2, kfuture = 1, local = !T),
0.305958, tolerance = 1e-6)
expect_equal(predictive_info(c(1, 1, 0, 0, 1, 0, 0, 1),
kpast = 2, kfuture = 2, local = !T),
1.521928, tolerance = 1e-6)
expect_equal(predictive_info(c(0, 0, 1, 1, 1, 1, 0, 0, 0),
kpast = 2, kfuture = 2, local = !T),
0.666667, tolerance = 1e-6)
expect_equal(predictive_info(c(1, 1, 0, 0, 1, 0, 0, 1),
kpast = 2, kfuture = 3, local = !T),
1.500000, tolerance = 1e-6)
expect_equal(predictive_info(c(0, 0, 1, 1, 1, 1, 0, 0, 0),
kpast = 2, kfuture = 3, local = !T),
0.970951, tolerance = 1e-6)
expect_equal(predictive_info(c(1, 1, 0, 0, 1, 0, 0, 1),
kpast = 1, kfuture = 2, local = !T),
0.666667, tolerance = 1e-6)
expect_equal(predictive_info(c(0, 0, 1, 1, 1, 1, 0, 0, 0),
kpast = 1, kfuture = 2, local = !T),
0.305958, tolerance = 1e-6)
expect_equal(predictive_info(c(1, 1, 0, 0, 1, 0, 0, 1),
kpast = 3, kfuture = 3, local = !T),
1.584963, tolerance = 1e-6)
expect_equal(predictive_info(c(0, 0, 1, 1, 1, 1, 0, 0, 0),
kpast = 3, kfuture = 3, local = !T),
1.500000, tolerance = 1e-6)
expect_equal(predictive_info(c(3, 3, 3, 2, 1, 0, 0, 0, 1),
kpast = 2, kfuture = 1, local = !T),
1.270942, tolerance = 1e-6)
expect_equal(predictive_info(c(2, 2, 2, 2, 2, 2, 1, 1, 1),
kpast = 2, kfuture = 1, local = !T),
0.469566, tolerance = 1e-6)
expect_equal(predictive_info(c(3, 3, 3, 2, 1, 0, 0, 0, 1),
kpast = 2, kfuture = 2, local = !T),
1.918296, tolerance = 1e-6)
expect_equal(predictive_info(c(2, 2, 2, 2, 2, 2, 1, 1, 1),
kpast = 2, kfuture = 2, local = !T),
0.316689, tolerance = 1e-6)
})
test_that("predictive_info on ensemble of series", {
series <- matrix(0, nrow = 8, ncol = 2)
series[, 1] <- c(1, 1, 0, 0, 1, 0, 0, 1)
series[, 2] <- c(0, 0, 0, 1, 0, 0, 0, 1)
expect_equal(predictive_info(series, kpast = 2, kfuture = 1, local = !T),
0.459148, tolerance = 1e-6)
expect_equal(predictive_info(series, kpast = 2, kfuture = 3, local = !T),
1.061278, tolerance = 1e-6)
series <- matrix(0, nrow = 9, ncol = 9)
series[, 1] <- c(1, 0, 0, 0, 0, 0, 0, 0, 0)
series[, 2] <- c(0, 0, 1, 1, 1, 1, 0, 0, 0)
series[, 3] <- c(1, 0, 0, 0, 0, 0, 0, 1, 1)
series[, 4] <- c(1, 0, 0, 0, 0, 0, 0, 1, 1)
series[, 5] <- c(0, 0, 0, 0, 0, 1, 1, 0, 0)
series[, 6] <- c(0, 0, 0, 0, 1, 1, 0, 0, 0)
series[, 7] <- c(1, 1, 1, 0, 0, 0, 0, 1, 1)
series[, 8] <- c(0, 0, 0, 1, 1, 1, 1, 0, 0)
series[, 9] <- c(0, 0, 0, 0, 0, 0, 1, 1, 0)
expect_equal(predictive_info(series, kpast = 2, kfuture = 1, local = !T),
0.3080467, tolerance = 1e-6)
expect_equal(predictive_info(series, kpast = 2, kfuture = 3, local = !T),
0.628720, tolerance = 1e-6)
series <- matrix(0, nrow = 9, ncol = 4)
series[, 1] <- c(3, 3, 3, 2, 1, 0, 0, 0, 1)
series[, 2] <- c(2, 2, 3, 3, 3, 3, 2, 1, 0)
series[, 3] <- c(0, 0, 0, 0, 1, 1, 0, 0, 0)
series[, 4] <- c(1, 1, 0, 0, 0, 1, 1, 2, 2)
expect_equal(predictive_info(series, kpast = 2, kfuture = 1, local = !T),
1.324292, tolerance = 1e-6)
expect_equal(predictive_info(series, kpast = 2, kfuture = 3, local = !T),
2.385475, tolerance = 1e-6)
})
test_that("predictive_info local on single series", {
expect_equal(mean(predictive_info(c(1, 1, 0, 0, 1, 0, 0, 1),
kpast = 2, kfuture = 1, local = T)),
0.918296, tolerance = 1e-6)
expect_equal(mean(predictive_info(c(0, 0, 1, 1, 1, 1, 0, 0, 0),
kpast = 2, kfuture = 1, local = T)),
0.305958, tolerance = 1e-6)
expect_equal(mean(predictive_info(c(1, 1, 0, 0, 1, 0, 0, 1),
kpast = 2, kfuture = 2, local = T)),
1.521928, tolerance = 1e-6)
expect_equal(mean(predictive_info(c(0, 0, 1, 1, 1, 1, 0, 0, 0),
kpast = 2, kfuture = 2, local = T)),
0.666667, tolerance = 1e-6)
expect_equal(mean(predictive_info(c(1, 1, 0, 0, 1, 0, 0, 1),
kpast = 2, kfuture = 3, local = T)),
1.500000, tolerance = 1e-6)
expect_equal(mean(predictive_info(c(0, 0, 1, 1, 1, 1, 0, 0, 0),
kpast = 2, kfuture = 3, local = T)),
0.970951, tolerance = 1e-6)
expect_equal(mean(predictive_info(c(1, 1, 0, 0, 1, 0, 0, 1),
kpast = 1, kfuture = 2, local = T)),
0.666667, tolerance = 1e-6)
expect_equal(mean(predictive_info(c(0, 0, 1, 1, 1, 1, 0, 0, 0),
kpast = 1, kfuture = 2, local = T)),
0.305958, tolerance = 1e-6)
expect_equal(mean(predictive_info(c(1, 1, 0, 0, 1, 0, 0, 1),
kpast = 3, kfuture = 3, local = T)),
1.584963, tolerance = 1e-6)
expect_equal(mean(predictive_info(c(0, 0, 1, 1, 1, 1, 0, 0, 0),
kpast = 3, kfuture = 3, local = T)),
1.500000, tolerance = 1e-6)
expect_equal(mean(predictive_info(c(3, 3, 3, 2, 1, 0, 0, 0, 1),
kpast = 2, kfuture = 1, local = T)),
1.270942, tolerance = 1e-6)
expect_equal(mean(predictive_info(c(2, 2, 2, 2, 2, 2, 1, 1, 1),
kpast = 2, kfuture = 1, local = T)),
0.469566, tolerance = 1e-6)
expect_equal(mean(predictive_info(c(3, 3, 3, 2, 1, 0, 0, 0, 1),
kpast = 2, kfuture = 2, local = T)),
1.918296, tolerance = 1e-6)
expect_equal(mean(predictive_info(c(2, 2, 2, 2, 2, 2, 1, 1, 1),
kpast = 2, kfuture = 2, local = T)),
0.316689, tolerance = 1e-6)
})
test_that("predictive_info local on ensemble of series", {
series <- matrix(0, nrow = 8, ncol = 2)
series[, 1] <- c(1, 1, 0, 0, 1, 0, 0, 1)
series[, 2] <- c(0, 0, 0, 1, 0, 0, 0, 1)
expect_equal(mean(predictive_info(series, kpast = 2, kfuture = 1, local = T)),
0.459148, tolerance = 1e-6)
expect_equal(mean(predictive_info(series, kpast = 2, kfuture = 3, local = T)),
1.061278, tolerance = 1e-6)
series <- matrix(0, nrow = 9, ncol = 9)
series[, 1] <- c(1, 0, 0, 0, 0, 0, 0, 0, 0)
series[, 2] <- c(0, 0, 1, 1, 1, 1, 0, 0, 0)
series[, 3] <- c(1, 0, 0, 0, 0, 0, 0, 1, 1)
series[, 4] <- c(1, 0, 0, 0, 0, 0, 0, 1, 1)
series[, 5] <- c(0, 0, 0, 0, 0, 1, 1, 0, 0)
series[, 6] <- c(0, 0, 0, 0, 1, 1, 0, 0, 0)
series[, 7] <- c(1, 1, 1, 0, 0, 0, 0, 1, 1)
series[, 8] <- c(0, 0, 0, 1, 1, 1, 1, 0, 0)
series[, 9] <- c(0, 0, 0, 0, 0, 0, 1, 1, 0)
expect_equal(mean(predictive_info(series, kpast = 2, kfuture = 1, local = T)),
0.3080467, tolerance = 1e-6)
expect_equal(mean(predictive_info(series, kpast = 2, kfuture = 3, local = T)),
0.628720, tolerance = 1e-6)
series <- matrix(0, nrow = 9, ncol = 4)
series[, 1] <- c(3, 3, 3, 2, 1, 0, 0, 0, 1)
series[, 2] <- c(2, 2, 3, 3, 3, 3, 2, 1, 0)
series[, 3] <- c(0, 0, 0, 0, 1, 1, 0, 0, 0)
series[, 4] <- c(1, 1, 0, 0, 0, 1, 1, 2, 2)
expect_equal(mean(predictive_info(series, kpast = 2, kfuture = 1, local = T)),
1.324292, tolerance = 1e-6)
expect_equal(mean(predictive_info(series, kpast = 2, kfuture = 3, local = T)),
2.385475, tolerance = 1e-6)
})
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.