Skip to contents

Estimate bias using t-test and augmented t-test

Source: R/tests.R
bias.Rd

Estimate bias using t-test and augmented t-test

Usage

bias(revisions.view, na.zero = FALSE)

Arguments

revisions.view

mts object. Vertical or diagonal view of the get_revisions() output

na.zero

Boolean whether missing values should be considered as 0 or rather as data not (yet) available (the default).

See also

revision_analysis(), render_report()

Examples


## Simulated data
df_long <- simulate_long(
    n_period = 10L * 4L,
    n_revision = 5L,
    periodicity = 4L,
    start_period = as.Date("2010-01-01")
)

## Create vintage and test
vintages <- create_vintages(df_long, periodicity = 4)
revisions <- get_revisions(vintages, gap = 1)
bias(revisions$diagonal_view)
#>                            N    estimate     stderr      tstat     pvalue
#> [Release[2]]-[Release[1]] 24 -0.54510403 0.49669780 -1.0974561 0.28379928
#> [Release[3]]-[Release[2]] 16  0.23607525 0.36069152  0.6545074 0.52269481
#> [Release[4]]-[Release[3]] 14 -0.25693314 0.12186524 -2.1083381 0.05497388
#> [Release[5]]-[Release[4]]  6  0.07740432 0.06517643  1.1876121 0.28832326
#>                                ar(1) stderr.adjusted tstat.adjusted
#> [Release[2]]-[Release[1]] -0.1498687      0.42708187     -1.2763455
#> [Release[3]]-[Release[2]]  0.4195115      0.56403825      0.4185448
#> [Release[4]]-[Release[3]]  0.3617668      0.17800884     -1.4433730
#> [Release[5]]-[Release[4]] -0.6275096      0.03118073      2.4824407
#>                           pvalue.adjusted
#> [Release[2]]-[Release[1]]      0.21088964
#> [Release[3]]-[Release[2]]      0.68894287
#> [Release[4]]-[Release[3]]      0.19490637
#> [Release[5]]-[Release[4]]      0.01976918