tests/testthat/test_2_repeatsales.R
In andykrause/hpiR: House Price Indexes
#*****************************************************************************************
# *
# Unit tests for hpiR package - Repeat Sales Functions *
# *
#*****************************************************************************************
library(hpiR)
library(testthat)
## Load Data
sales <- get(data(seattle_sales))
### rtCreateTrans() ----------------------------------------------------------------------
context('rtCreateTrans()')
# Test Setup
test_that("Can take a functional 'trans_df' object", {
sales_df <- dateToPeriod(trans_df = sales,
date = 'sale_date',
periodicity = 'monthly')
expect_is(rt_df <- rtCreateTrans(trans_df=sales_df,
prop_id='pinx',
trans_id='sale_id',
price='sale_price'),
'rtdata')
expect_true(nrow(rt_df) == 5102)
})
# Test Setup
test_that("Can create own salesdf object", {
expect_is(rt_df <- rtCreateTrans(trans_df=sales,
prop_id='pinx',
trans_id='sale_id',
price='sale_price',
date='sale_date',
periodicity='monthly'),
'rtdata')
expect_true(nrow(rt_df) == 5102)
})
test_that("Can use min/max dates own salesdf object", {
expect_is(rt_df <- rtCreateTrans(trans_df=sales,
prop_id='pinx',
trans_id='sale_id',
price='sale_price',
date='sale_date',
periodicity='monthly',
min_date = as.Date('2012-03-21')),
'rtdata')
expect_true(nrow(rt_df) == 5102)
expect_is(rt_df <- rtCreateTrans(trans_df=sales,
prop_id='pinx',
trans_id='sale_id',
price='sale_price',
date='sale_date',
periodicity='monthly',
min_date = as.Date('2012-03-21'),
adj_type='clip'),
'rtdata')
expect_true(nrow(rt_df) == 2827)
expect_is(rt_df <- rtCreateTrans(trans_df=sales,
prop_id='pinx',
trans_id='sale_id',
price='sale_price',
date='sale_date',
periodicity='monthly',
max_date = as.Date('2015-03-21')),
'rtdata')
expect_true(nrow(rt_df) == 5102)
expect_is(rt_df <- rtCreateTrans(trans_df=sales,
prop_id='pinx',
trans_id='sale_id',
price='sale_price',
date='sale_date',
periodicity='monthly',
max_date = as.Date('2014-03-21'),
adj_type='clip'),
'rtdata')
expect_true(nrow(rt_df) == 1148)
})
test_that("Sequence only (seq_only) option works", {
expect_is(rt_df <- rtCreateTrans(trans_df=sales,
prop_id='pinx',
trans_id='sale_id',
price='sale_price',
date='sale_date',
periodicity='monthly',
seq_only = TRUE),
'rtdata')
expect_true(nrow(rt_df) == 4823)
})
test_that("min_period_dist argument works", {
expect_is(rt_df <- rtCreateTrans(trans_df=sales,
prop_id='pinx',
trans_id='sale_id',
price='sale_price',
date='sale_date',
periodicity='monthly',
seq_only = TRUE,
min_period_dist = 12),
'rtdata')
expect_true(nrow(rt_df) == 3795)
})
test_that("Fails if sales creation fails", {
# Bad Date field
expect_error(rt_df <- rtCreateTrans(trans_df=sales,
prop_id='pinx',
trans_id='sale_id',
price='sale_price',
date='sale_price',
periodicity='monthly'))
# Bad Periodicity field
expect_error(rt_df <- rtCreateTrans(trans_df=sales,
prop_id='pinx',
trans_id='sale_id',
price='sale_price',
date='sale_date',
periodicity='mocnthly'))
})
# Create a sales_df in the global env for future use
sales_df <- dateToPeriod(trans_df = sales,
date = 'sale_date',
periodicity = 'monthly')
test_that("Fails if bad arguments fails", {
# Bad prop_id field
expect_error(rt_df <- rtCreateTrans(trans_df=sales_df,
prop_id='pinxx',
trans_id='sale_id',
price='sale_price'))
# Bad sale_id field
expect_error(rt_df <- rtCreateTrans(trans_df=sales_df,
prop_id='pinx',
trans_id='salex_id',
price='sale_price'))
# Bad price field
expect_error(rt_df <- rtCreateTrans(trans_df=sales_df,
prop_id='pinx',
trans_id='sale_id',
price='salex_price'))
})
test_that("Returns NULL if no repeat sales", {
expect_is(rt_df <- rtCreateTrans(trans_df=sales_df[!duplicated(sales_df$prop_id),],
prop_id='pinx',
trans_id='sale_id',
price='sale_price'), "NULL")
expect_is(rt_df <- rtCreateTrans(trans_df=sales_df[1:3, ],
prop_id='pinx',
trans_id='sale_id',
price='sale_price'), "NULL")
})
### rtTimeMatrix -------------------------------------------------------------------------
context('rtTimeMatrix()')
# Create a full rt_data object in global environment
rt_df <- rtCreateTrans(trans_df=sales_df,
prop_id='pinx',
trans_id='sale_id',
price='sale_price')
test_that('Time matrix operates properly', {
# Work with an 'rtdata' object
expect_is(time_matrix <- rtTimeMatrix(rt_df), 'timematrix')
# Returns a NULL without
expect_error(time_matrix <- rtTimeMatrix(sales_df))
# Returns correct number of rows
expect_true(nrow(rtTimeMatrix(rt_df[1:2000,])) == 2000)
# Return correct number of columns
expect_true(ncol(rtTimeMatrix(rt_df[1:2000,])) ==
nrow(attr(rt_df, 'period_table')) - 1)
})
## hpiModel.rtdata(): Prior to rtModel() call -------------------------------------------
context('hpiModel.rtdata(): Prior to rtModel() call')
test_that('hpiModel.rtdata works in simplest format',{
expect_is(rt_model <- hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'base',
log_dep = TRUE),
'hpimodel')
})
test_that('"log_dep" argument works both ways',{
# TRUE
expect_true(hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'base',
log_dep = TRUE)$model_obj$fitted.values[1] < 1)
# FALSE
expect_true(hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'base',
log_dep = FALSE)$model_obj$fitted.values[1] > 10000)
})
test_that('Check for zero or negative prices works',{
rt_dfx <- rt_df
# Zero Price
rt_dfx$price_1[1] <- 0
expect_error(rt_model <- hpiModel(model_type = 'rt',
hpi_df = rt_dfx,
estimator = 'base',
log_dep = TRUE))
expect_is(rt_model <- hpiModel(model_type = 'rt',
hpi_df = rt_dfx,
estimator = 'base',
log_dep = FALSE),
'hpimodel')
# NA
rt_dfx$price_1[1] <- NA_integer_
expect_error(rt_model <- hpiModel(model_type = 'rt',
hpi_df = rt_dfx,
estimator = 'base',
log_dep = TRUE))
expect_error(rt_model <- hpiModel(model_type = 'rt',
hpi_df = rt_dfx,
estimator = 'base',
log_dep = FALSE))
# Infinity
rt_dfx$price_1[1] <- Inf
expect_error(rt_model <- hpiModel(model_type = 'rt',
hpi_df = rt_dfx,
estimator = 'base',
log_dep = TRUE))
expect_error(rt_model <- hpiModel(model_type = 'rt',
hpi_df = rt_dfx,
estimator = 'base',
log_dep = FALSE))
})
test_that('Check for estimator type works',{
# Base
expect_true(hpiModel(model_type = 'rt',
hpi_df = rt_df)$estimator == 'base')
# Convert to base with bad estimator
expect_true(hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator='xxxx')$estimator == 'base')
# Robust
expect_true(hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator='robust')$estimator == 'robust')
# Weighted
expect_true(hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator='weighted')$estimator == 'weighted')
})
## rtModel() ------------------------------------------------------------------------
context('rtModel()')
# Create matrix and difference vectort in global environment
time_matrix <- rtTimeMatrix(rt_df)
price_diff_l <- log(rt_df$price_2) - log(rt_df$price_1)
price_diff <- rt_df$price_2 - rt_df$price_1
test_that('Check for errort with bad arguments',{
# Base: Return warning if wrong rt_df
expect_error(rt_model <- rtModel(rt_df = sales,
time_matrix = time_matrix,
price_diff = price_diff_l,
estimator=structure('base', class='base')))
# Robust: Return warning if wrong time_matrix
expect_error(rt_model <- rtModel(rt_df = rt_df,
time_matrix = sales,
price_diff = price_diff_l,
estimator=structure('robust', class='robust')))
# Weighted: Dimensions of data do not match
expect_error(rt_model <- rtModel(rt_df = rt_df,
time_matrix = time_matrix,
price_diff = price_diff_l[-1],
estimator=structure('weighted', class='weighted')))
# Bad estimator class
expect_error(rt_model <- rtModel(rt_df = rt_df,
time_matrix = time_matrix,
price_diff = price_diff,
estimator=structure('base', class='xxx')))
})
test_that('Performance with sparse data',{
## Moderate Sparseness
# Create a sparse data set
rt_df200 <- rt_df[1:200, ]
time_matrix200 <- rtTimeMatrix(rt_df200)
price_diff_l200 <- log(rt_df200$price_2) - log(rt_df200$price_1)
price_diff200 <- rt_df200$price_2 - rt_df200$price_1
# Works with base, though many NAs
expect_is(rt_model <- rtModel(rt_df = rt_df200,
time_matrix = time_matrix200,
price_diff = price_diff_l200,
estimator=structure('base', class='base')),
'rtmodel')
# Robust works but gives warning
expect_is(rt_model <- rtModel(rt_df = rt_df200,
time_matrix = time_matrix200,
price_diff = price_diff_l200,
estimator=structure('robust', class='robust')),
'rtmodel')
# Weighted works
expect_is(rt_model <- rtModel(rt_df = rt_df200,
time_matrix = time_matrix200,
price_diff = price_diff200,
estimator=structure('weighted', class='weighted')),
'rtmodel')
## Check severe sparteness
# Create data set
rt_df20 <- rt_df[1:20, ]
time_matrix20 <- rtTimeMatrix(rt_df20)
price_diff_l20 <- log(rt_df20$price_2) - log(rt_df20$price_1)
price_diff20 <- rt_df20$price_2 - rt_df20$price_1
# Base: Works, but gives message
expect_is(rt_model <- rtModel(rt_df = rt_df20,
time_matrix = time_matrix20,
price_diff = price_diff_l20,
estimator=structure('base', class='base')),
'rtmodel')
# Robust: Works, but gives warning from lmrob()
expect_warning(rt_model <- rtModel(rt_df = rt_df20,
time_matrix = time_matrix20,
price_diff = price_diff_l20,
estimator=structure('robust', class='robust')))
# Weighted: works but gives message
expect_is(rt_model <- rtModel(rt_df = rt_df20,
time_matrix = time_matrix20,
price_diff = price_diff20,
estimator=structure('weighted', class='weighted')),
'rtmodel')
})
## hpiModel.rtdata() after rtModel --------------------------------------------------------
context('hpiModel.rtdata(): after rtModel()')
test_that('hpiModel.rtdata works in both trim_model cases', {
expect_is(rt_model <- hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'base',
log_dep = TRUE,
trim_model=TRUE), 'hpimodel')
expect_is(rt_model <- hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'base',
log_dep = TRUE,
trim_model=FALSE), 'hpimodel')
expect_is(rt_model <- hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'base',
log_dep = FALSE,
trim_model=TRUE), 'hpimodel')
expect_is(rt_model <- hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'base',
log_dep = FALSE,
trim_model=FALSE), 'hpimodel')
expect_is(rt_model <- hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'weighted',
log_dep = TRUE,
trim_model=FALSE), 'hpimodel')
expect_is(rt_model <- hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'robust',
log_dep = FALSE,
trim_model=TRUE), 'hpimodel')
expect_true(is.null(hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'weighted',
log_dep = TRUE,
trim_model=TRUE)$model_obj$qr))
expect_true(!is.null(hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'weighted',
log_dep = TRUE,
trim_model=FALSE)$model_obj$qr))
})
test_that('hpiModel.rtdata outputs are correct', {
# Run a model of each estimator type
rt_model_base <- hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'base',
log_dep = TRUE,
trim_model=TRUE)
rt_model_robust <- hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'robust',
log_dep = TRUE,
trim_model=FALSE)
rt_model_wgt <- hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'weighted',
log_dep = FALSE,
trim_model=TRUE)
set.seed(123)
rt_model_wwgt <- hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'weighted',
log_dep = FALSE,
weights = runif(nrow(rt_df), 0, 1),
trim_model=TRUE)
# Estimatort
expect_is(rt_model_base$estimator, 'base')
expect_is(rt_model_robust$estimator, 'robust')
expect_is(rt_model_wgt$estimator, 'weighted')
# Coefficients
expect_is(rt_model_base$coefficients, 'data.frame')
expect_is(rt_model_robust$coefficients, 'data.frame')
expect_is(rt_model_wgt$coefficients, 'data.frame')
expect_true(nrow(rt_model_base$coefficients) == 84)
expect_true(max(rt_model_robust$coefficients$time) == 84)
expect_true(rt_model_wgt$coefficients$coefficient[1] == 0)
expect_false(identical(
rt_model_wgt$coefficients$coefficient,
rt_model_wwgt$coefficients$coefficient))
# Modelobj
expect_is(rt_model_base$model_obj, 'rtmodel')
expect_is(rt_model_robust$model_obj, 'rtmodel')
expect_is(rt_model_wgt$model_obj, 'rtmodel')
# Model spec
expect_true(is.null(rt_model_base$model_spec))
expect_true(is.null(rt_model_robust$model_spec))
expect_true(is.null(rt_model_wgt$model_spec))
# base price
expect_true(round(rt_model_base$base_price, 0) == 427785)
expect_true(round(rt_model_robust$base_price, 0) == 427785)
expect_true(round(rt_model_wgt$base_price, 0) == 427785)
# Periods
expect_is(rt_model_base$periods, 'data.frame')
expect_true(nrow(rt_model_base$periods) == 84)
expect_is(rt_model_robust$periods, 'data.frame')
expect_true(nrow(rt_model_robust$periods) == 84)
expect_is(rt_model_wgt$periods, 'data.frame')
expect_true(nrow(rt_model_wgt$periods) == 84)
# Approach
expect_true(rt_model_base$approach == 'rt')
expect_true(rt_model_robust$approach == 'rt')
expect_true(rt_model_wgt$approach == 'rt')
})
### modelToIndex --------------------------------------------------------------------
context('modelToIndex')
rt_model <- hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'base',
log_dep = TRUE,
trim_model=TRUE)
test_that('modelToIndex works', {
expect_is(modelToIndex(rt_model), 'hpiindex')
})
test_that('modelToIndex works with other estimatort and options', {
# Robust, LogDep=T, TrimModel=T
expect_is(modelToIndex(hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'robust',
log_dep = TRUE,
trim_model=TRUE)), 'hpiindex')
# Weighted, LogDep=T, TrimModel=T
expect_is(modelToIndex(hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'weighted',
log_dep = TRUE,
trim_model=TRUE)), 'hpiindex')
# Robust, LogDep=F, TrimModel=T
expect_is(modelToIndex(hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'robust',
log_dep = FALSE,
trim_model=TRUE)), 'hpiindex')
# Weighted, LogDep=T, TrimModel=F
expect_is(modelToIndex(hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'weighted',
log_dep = TRUE,
trim_model=FALSE)), 'hpiindex')
})
test_that('modelToIndex fails with an error',{
expect_error(modelToIndex(model_obj = 'abc'))
})
test_that('modelToIndex imputes properly, BASE model, LogDEP',{
model_base <- hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'base',
log_dep = TRUE,
trim_model=TRUE)
# Impute a beginning value
model_ex <- model_base
model_ex$coefficients$coefficient[2] <- NA_real_
expect_is(modelToIndex(model_ex), 'hpiindex')
expect_true(!is.na(modelToIndex(model_ex)$value[2]))
expect_true(modelToIndex(model_ex)$value[2] == 100)
expect_true(modelToIndex(model_ex)$imputed[2] == 1)
# Interpolate interior values
model_ex <- model_base
model_ex$coefficients$coefficient[3:5] <- NA_real_
expect_is(modelToIndex(model_ex), 'hpiindex')
expect_true(all(!is.na(modelToIndex(model_ex)$value[3:5])))
# Extrapolate end periods
model_ex <- model_base
model_ex$coefficients$coefficient[81:84] <- NA_real_
expect_is(modelToIndex(model_ex), 'hpiindex')
expect_true(all(!is.na(modelToIndex(model_ex)$value[81:84])))
expect_true(modelToIndex(model_ex)$value[80] ==
modelToIndex(model_ex)$value[84])
})
test_that('modelToIndex imputes properly, BASE model, LogDep=FALSE',{
model_base <- hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'base',
log_dep = FALSE,
trim_model=TRUE)
# Extrapolate a beginning value
model_ex <- model_base
model_ex$coefficients$coefficient[2] <- NA_real_
expect_is(modelToIndex(model_ex), 'hpiindex')
expect_true(!is.na(modelToIndex(model_ex)$value[2]))
expect_true(modelToIndex(model_ex)$value[2] == 100)
expect_true(modelToIndex(model_ex)$imputed[2] == 1)
# Impute interior values
model_ex <- model_base
model_ex$coefficients$coefficient[3:5] <- NA_real_
expect_is(modelToIndex(model_ex), 'hpiindex')
expect_true(all(!is.na(modelToIndex(model_ex)$value[3:5])))
# Extrapolate an end value
model_ex <- model_base
model_ex$coefficients$coefficient[81:84] <- NA_real_
expect_is(modelToIndex(model_ex), 'hpiindex')
expect_true(all(!is.na(modelToIndex(model_ex)$value[81:84])))
expect_true(modelToIndex(model_ex)$value[80] ==
modelToIndex(model_ex)$value[84])
})
test_that('modelToIndex imputes properly, Robust model, LogDEP = TRUE',{
model_base <- hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'robust',
log_dep = TRUE,
trim_model=TRUE)
model_ex <- model_base
model_ex$coefficients$coefficient[2] <- NA_real_
expect_is(modelToIndex(model_ex), 'hpiindex')
expect_true(!is.na(modelToIndex(model_ex)$value[2]))
expect_true(modelToIndex(model_ex)$value[2] == 100)
expect_true(modelToIndex(model_ex)$imputed[2] == 1)
model_ex <- model_base
model_ex$coefficients$coefficient[3:5] <- NA_real_
expect_is(modelToIndex(model_ex), 'hpiindex')
expect_true(all(!is.na(modelToIndex(model_ex)$value[3:5])))
model_ex <- model_base
model_ex$coefficients$coefficient[81:84] <- NA_real_
expect_is(modelToIndex(model_ex), 'hpiindex')
expect_true(all(!is.na(modelToIndex(model_ex)$value[81:84])))
expect_true(modelToIndex(model_ex)$value[80] ==
modelToIndex(model_ex)$value[84])
})
test_that('modelToIndex imputes properly, Weighted model, LogDep=FALSE',{
model_base <- hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'weighted',
log_dep = FALSE,
trim_model=TRUE)
model_ex <- model_base
model_ex$coefficients$coefficient[2] <- NA_real_
expect_is(modelToIndex(model_ex), 'hpiindex')
expect_true(!is.na(modelToIndex(model_ex)$value[2]))
expect_true(modelToIndex(model_ex)$value[2] == 100)
expect_true(modelToIndex(model_ex)$imputed[2] == 1)
model_ex <- model_base
model_ex$coefficients$coefficient[3:5] <- NA_real_
expect_is(modelToIndex(model_ex), 'hpiindex')
expect_true(all(!is.na(modelToIndex(model_ex)$value[3:5])))
model_ex <- model_base
model_ex$coefficients$coefficient[81:84] <- NA_real_
expect_is(modelToIndex(model_ex), 'hpiindex')
expect_true(all(!is.na(modelToIndex(model_ex)$value[81:84])))
expect_true(modelToIndex(model_ex)$value[80] ==
modelToIndex(model_ex)$value[84])
})
test_that('modelToIndex "max_period" cutoff works',{
model_base <- hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'weighted',
log_dep = FALSE,
trim_model=TRUE)
# Create shortened index
index_80 <- modelToIndex(model_base, max_period = 80)
expect_is(index_80, 'hpiindex')
expect_true(length(index_80$value) == 80)
})
### smoothIndex() -------------------------------------------------------------
context('smoothIndex()')
model_base <- hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'base',
log_dep = TRUE,
trim_model=TRUE)
index_base <- modelToIndex(model_obj = model_base)
test_that('smoothing Function works with a variety of inputs',{
# HPI Index object
expect_is(index_smooth <- smoothIndex(index_obj = index_base,
order = 4),
'indexsmooth')
})
test_that('Errors are given when index is bad',{
# Non-sensical index
expect_error(index_smooth <- smoothIndex(index_obj = 'abc',
order = 3))
# Negative Order
expect_error(index_smooth <- smoothIndex(index_obj = index_base,
order = -3))
# Char Window
expect_error(index_smooth <- smoothIndex(index_obj = index_base,
order = 'x'))
# NA Window
expect_error(index_smooth <- smoothIndex(index_obj = index_base,
order = NA_integer_))
})
test_that('Returning in place works',{
# Add it to the Full HPI Object (to the hpiindex object)
expect_is(index_base <- smoothIndex(index = index_base,
order = 3,
in_place = TRUE),
'hpiindex')
expect_true('smooth' %in% names(index_base))
})
### rtIndex() wrapper ---------------------------------------------------------------
context('rtindex() wrapper')
test_that('Function works with proper inputs',{
# Full case
full_1 <- rtIndex(trans_df = sales,
date = 'sale_date',
price = 'sale_price',
trans_id = 'sale_id',
prop_id = 'pinx',
estimator = 'base',
log_dep = TRUE,
periodicity = 'monthly')
expect_is(full_1, 'hpi')
expect_true(full_1$model$estimator == 'base')
# Giving a 'trans_df' object
full_2 <- rtIndex(trans_df = sales_df,
price = 'sale_price',
trans_id = 'sale_id',
prop_id = 'pinx',
estimator = 'robust',
log_dep = TRUE)
expect_is(full_2, 'hpi')
expect_true(full_2$model$estimator == 'robust')
# Giving an 'rt_df' object
full_3 <- rtIndex(trans_df = rt_df,
estimator = 'weighted',
log_dep = TRUE)
expect_is(full_3, 'hpi')
expect_true(full_3$model$estimator == 'weighted')
})
test_that('Additional arguments in rtIndex() work',{
## rt Create arguments
# Min Date Model with Clip
mindate_index <- rtIndex(trans_df = sales,
date = 'sale_date',
price = 'sale_price',
trans_id = 'sale_id',
prop_id = 'pinx',
min_date = as.Date('2011-01-01'),
adj_type = 'clip')
expect_true(mindate_index$index$name[1] == '2011')
# Max Date Model with Adjust
maxdate_index <- rtIndex(trans_df = sales,
date = 'sale_date',
price = 'sale_price',
trans_id = 'sale_id',
prop_id = 'pinx',
max_date = as.Date('2015-12-31'))
expect_true(maxdate_index$index$name[7] == '2016')
# Periodicity
per_index <- rtIndex(trans_df = sales,
date = 'sale_date',
price = 'sale_price',
trans_id = 'sale_id',
prop_id = 'pinx',
periodicity = 'weekly')
expect_true(per_index$index$name[364] == 'week: 2016-12-11 to 2016-12-17')
# Sequence Only
seq_index <- rtIndex(trans_df = sales_df,
date = 'sale_date',
price = 'sale_price',
trans_id = 'sale_id',
prop_id = 'pinx',
seq_only = TRUE)
expect_true(nrow(seq_index$data) == 4823)
## HPI Model
# Trim Model
trim_index <- rtIndex(trans_df = rt_df,
trim_model=TRUE)
expect_true(is.null(trim_index$model$model_obj$qr))
# Log Dep & Robust
ld_index <- rtIndex(trans_df = rt_df,
estimator = 'robust',
log_dep = FALSE)
expect_true(ld_index$model$log_dep == FALSE)
expect_true(ld_index$model$estimator == 'robust')
## Model to Index
m2i_index <- rtIndex(trans_df = rt_df,
estimator = 'robust',
log_dep = FALSE,
max_period = 80)
expect_true(length(m2i_index$index$value) == 80)
})
test_that("Bad arguments generate Errors: Full Case",{
expect_error(rtIndex(trans_df = sales,
date = 'sale_price',
price = 'sale_price',
trans_id = 'sale_id',
prop_id = 'pinx',
estimator = 'base',
log_dep = TRUE,
periodicity = 'monthly'))
expect_error(rtIndex(trans_df = sales,
date = 'sale_date',
price = 'sale_price',
prop_id = 'pinx',
estimator = 'base',
log_dep = TRUE,
periodicity = 'monthly'))
expect_error(rtIndex(trans_df = sales,
date = 'sale_date',
price = 'sale_price',
trans_id = 'sale_id',
estimator = 'base',
log_dep = TRUE,
periodicity = 'monthly'))
expect_error(rtIndex(trans_df = sales,
date = 'sale_date',
trans_id = 'sale_id',
prop_id = 'pinx',
estimator = 'base',
log_dep = TRUE,
periodicity = 'monthly'))
expect_error(rtIndex(trans_df = sales,
date = 'sale_date',
price = 'sale_price',
trans_id = 'sale_id',
prop_id = 'pinx',
estimator = 'base',
log_dep = TRUE,
periodicity = 'xxx'))
expect_error(rtIndex(trans_df = sales[1, ],
date = 'sale_date',
price = 'sale_price',
trans_id = 'sale_id',
prop_id = 'pinx',
estimator = 'base',
log_dep = TRUE,
periodicity = 'monthly'), 'Converting transactions')
})
test_that("Bad arguments generate errort: trans_df Case",{
expect_error(rtIndex(trans_df = sales_df,
price = 'xx',
trans_id = 'sale_id',
prop_id = 'pinx',
estimator = 'base',
log_dep = TRUE))
expect_error(rtIndex(trans_df = sales_df,
price = 'sale_price',
trans_id = 'xx',
prop_id = 'pinx',
estimator = 'base',
log_dep = TRUE))
expect_error(rtIndex(trans_df = sales_df,
price = 'sale_price',
prop_id = 'pinx',
estimator = 'base',
log_dep = TRUE))
expect_error(rtIndex(trans_df = sales_df,
price = 'sale_price',
trans_id = 'sale_id',
prop_id = 'xx',
estimator = 'base',
log_dep = TRUE))
})
test_that("Bad arguments handling: rt_sales Case",{
# Bad estimatort default to 'base'
expect_true(rtIndex(trans_df = rt_df,
estimator = 'basex',
log_dep = TRUE)$model$estimator == 'base')
expect_error(rtIndex(trans_df = rt_df,
estimator = 'robust',
log_dep = 'a'))
expect_error(rtIndex(trans_df = rt_df,
estimator = 'robust',
trim_model = 'a'))
expect_error(rtIndex(trans_df = rt_df,
estimator = 'robust',
max_period = 'a'))
})
test_that("Smoothing in_place for 'hpi' object works",{
full_1 <- rtIndex(trans_df = sales,
date = 'sale_date',
price = 'sale_price',
trans_id = 'sale_id',
prop_id = 'pinx',
estimator = 'base',
log_dep = TRUE,
periodicity = 'monthly',
smooth = TRUE)
# In wrapper smoothing worked
expect_is(full_1$index$smooth, 'indexsmooth')
# Full HPI Object
expect_is(index_smooth <- smoothIndex(index_obj = full_1,
order = 6),
'indexsmooth')
# Add it to the Full HPI Object (to the hpiindex object) with new name
expect_is(full_1s <- smoothIndex(index = full_1,
order = 3,
in_place = TRUE),
'hpi')
expect_is(full_1s$index$smooth, 'ts')
expect_is(full_1s$index$smooth, 'indexsmooth')
})
#*****************************************************************************************
#*****************************************************************************************
andykrause/hpiR documentation built on Feb. 2, 2024, 8:13 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
#*****************************************************************************************
# *
# Unit tests for hpiR package - Repeat Sales Functions *
# *
#*****************************************************************************************
library(hpiR)
library(testthat)
## Load Data
sales <- get(data(seattle_sales))
### rtCreateTrans() ----------------------------------------------------------------------
context('rtCreateTrans()')
# Test Setup
test_that("Can take a functional 'trans_df' object", {
sales_df <- dateToPeriod(trans_df = sales,
date = 'sale_date',
periodicity = 'monthly')
expect_is(rt_df <- rtCreateTrans(trans_df=sales_df,
prop_id='pinx',
trans_id='sale_id',
price='sale_price'),
'rtdata')
expect_true(nrow(rt_df) == 5102)
})
# Test Setup
test_that("Can create own salesdf object", {
expect_is(rt_df <- rtCreateTrans(trans_df=sales,
prop_id='pinx',
trans_id='sale_id',
price='sale_price',
date='sale_date',
periodicity='monthly'),
'rtdata')
expect_true(nrow(rt_df) == 5102)
})
test_that("Can use min/max dates own salesdf object", {
expect_is(rt_df <- rtCreateTrans(trans_df=sales,
prop_id='pinx',
trans_id='sale_id',
price='sale_price',
date='sale_date',
periodicity='monthly',
min_date = as.Date('2012-03-21')),
'rtdata')
expect_true(nrow(rt_df) == 5102)
expect_is(rt_df <- rtCreateTrans(trans_df=sales,
prop_id='pinx',
trans_id='sale_id',
price='sale_price',
date='sale_date',
periodicity='monthly',
min_date = as.Date('2012-03-21'),
adj_type='clip'),
'rtdata')
expect_true(nrow(rt_df) == 2827)
expect_is(rt_df <- rtCreateTrans(trans_df=sales,
prop_id='pinx',
trans_id='sale_id',
price='sale_price',
date='sale_date',
periodicity='monthly',
max_date = as.Date('2015-03-21')),
'rtdata')
expect_true(nrow(rt_df) == 5102)
expect_is(rt_df <- rtCreateTrans(trans_df=sales,
prop_id='pinx',
trans_id='sale_id',
price='sale_price',
date='sale_date',
periodicity='monthly',
max_date = as.Date('2014-03-21'),
adj_type='clip'),
'rtdata')
expect_true(nrow(rt_df) == 1148)
})
test_that("Sequence only (seq_only) option works", {
expect_is(rt_df <- rtCreateTrans(trans_df=sales,
prop_id='pinx',
trans_id='sale_id',
price='sale_price',
date='sale_date',
periodicity='monthly',
seq_only = TRUE),
'rtdata')
expect_true(nrow(rt_df) == 4823)
})
test_that("min_period_dist argument works", {
expect_is(rt_df <- rtCreateTrans(trans_df=sales,
prop_id='pinx',
trans_id='sale_id',
price='sale_price',
date='sale_date',
periodicity='monthly',
seq_only = TRUE,
min_period_dist = 12),
'rtdata')
expect_true(nrow(rt_df) == 3795)
})
test_that("Fails if sales creation fails", {
# Bad Date field
expect_error(rt_df <- rtCreateTrans(trans_df=sales,
prop_id='pinx',
trans_id='sale_id',
price='sale_price',
date='sale_price',
periodicity='monthly'))
# Bad Periodicity field
expect_error(rt_df <- rtCreateTrans(trans_df=sales,
prop_id='pinx',
trans_id='sale_id',
price='sale_price',
date='sale_date',
periodicity='mocnthly'))
})
# Create a sales_df in the global env for future use
sales_df <- dateToPeriod(trans_df = sales,
date = 'sale_date',
periodicity = 'monthly')
test_that("Fails if bad arguments fails", {
# Bad prop_id field
expect_error(rt_df <- rtCreateTrans(trans_df=sales_df,
prop_id='pinxx',
trans_id='sale_id',
price='sale_price'))
# Bad sale_id field
expect_error(rt_df <- rtCreateTrans(trans_df=sales_df,
prop_id='pinx',
trans_id='salex_id',
price='sale_price'))
# Bad price field
expect_error(rt_df <- rtCreateTrans(trans_df=sales_df,
prop_id='pinx',
trans_id='sale_id',
price='salex_price'))
})
test_that("Returns NULL if no repeat sales", {
expect_is(rt_df <- rtCreateTrans(trans_df=sales_df[!duplicated(sales_df$prop_id),],
prop_id='pinx',
trans_id='sale_id',
price='sale_price'), "NULL")
expect_is(rt_df <- rtCreateTrans(trans_df=sales_df[1:3, ],
prop_id='pinx',
trans_id='sale_id',
price='sale_price'), "NULL")
})
### rtTimeMatrix -------------------------------------------------------------------------
context('rtTimeMatrix()')
# Create a full rt_data object in global environment
rt_df <- rtCreateTrans(trans_df=sales_df,
prop_id='pinx',
trans_id='sale_id',
price='sale_price')
test_that('Time matrix operates properly', {
# Work with an 'rtdata' object
expect_is(time_matrix <- rtTimeMatrix(rt_df), 'timematrix')
# Returns a NULL without
expect_error(time_matrix <- rtTimeMatrix(sales_df))
# Returns correct number of rows
expect_true(nrow(rtTimeMatrix(rt_df[1:2000,])) == 2000)
# Return correct number of columns
expect_true(ncol(rtTimeMatrix(rt_df[1:2000,])) ==
nrow(attr(rt_df, 'period_table')) - 1)
})
## hpiModel.rtdata(): Prior to rtModel() call -------------------------------------------
context('hpiModel.rtdata(): Prior to rtModel() call')
test_that('hpiModel.rtdata works in simplest format',{
expect_is(rt_model <- hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'base',
log_dep = TRUE),
'hpimodel')
})
test_that('"log_dep" argument works both ways',{
# TRUE
expect_true(hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'base',
log_dep = TRUE)$model_obj$fitted.values[1] < 1)
# FALSE
expect_true(hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'base',
log_dep = FALSE)$model_obj$fitted.values[1] > 10000)
})
test_that('Check for zero or negative prices works',{
rt_dfx <- rt_df
# Zero Price
rt_dfx$price_1[1] <- 0
expect_error(rt_model <- hpiModel(model_type = 'rt',
hpi_df = rt_dfx,
estimator = 'base',
log_dep = TRUE))
expect_is(rt_model <- hpiModel(model_type = 'rt',
hpi_df = rt_dfx,
estimator = 'base',
log_dep = FALSE),
'hpimodel')
# NA
rt_dfx$price_1[1] <- NA_integer_
expect_error(rt_model <- hpiModel(model_type = 'rt',
hpi_df = rt_dfx,
estimator = 'base',
log_dep = TRUE))
expect_error(rt_model <- hpiModel(model_type = 'rt',
hpi_df = rt_dfx,
estimator = 'base',
log_dep = FALSE))
# Infinity
rt_dfx$price_1[1] <- Inf
expect_error(rt_model <- hpiModel(model_type = 'rt',
hpi_df = rt_dfx,
estimator = 'base',
log_dep = TRUE))
expect_error(rt_model <- hpiModel(model_type = 'rt',
hpi_df = rt_dfx,
estimator = 'base',
log_dep = FALSE))
})
test_that('Check for estimator type works',{
# Base
expect_true(hpiModel(model_type = 'rt',
hpi_df = rt_df)$estimator == 'base')
# Convert to base with bad estimator
expect_true(hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator='xxxx')$estimator == 'base')
# Robust
expect_true(hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator='robust')$estimator == 'robust')
# Weighted
expect_true(hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator='weighted')$estimator == 'weighted')
})
## rtModel() ------------------------------------------------------------------------
context('rtModel()')
# Create matrix and difference vectort in global environment
time_matrix <- rtTimeMatrix(rt_df)
price_diff_l <- log(rt_df$price_2) - log(rt_df$price_1)
price_diff <- rt_df$price_2 - rt_df$price_1
test_that('Check for errort with bad arguments',{
# Base: Return warning if wrong rt_df
expect_error(rt_model <- rtModel(rt_df = sales,
time_matrix = time_matrix,
price_diff = price_diff_l,
estimator=structure('base', class='base')))
# Robust: Return warning if wrong time_matrix
expect_error(rt_model <- rtModel(rt_df = rt_df,
time_matrix = sales,
price_diff = price_diff_l,
estimator=structure('robust', class='robust')))
# Weighted: Dimensions of data do not match
expect_error(rt_model <- rtModel(rt_df = rt_df,
time_matrix = time_matrix,
price_diff = price_diff_l[-1],
estimator=structure('weighted', class='weighted')))
# Bad estimator class
expect_error(rt_model <- rtModel(rt_df = rt_df,
time_matrix = time_matrix,
price_diff = price_diff,
estimator=structure('base', class='xxx')))
})
test_that('Performance with sparse data',{
## Moderate Sparseness
# Create a sparse data set
rt_df200 <- rt_df[1:200, ]
time_matrix200 <- rtTimeMatrix(rt_df200)
price_diff_l200 <- log(rt_df200$price_2) - log(rt_df200$price_1)
price_diff200 <- rt_df200$price_2 - rt_df200$price_1
# Works with base, though many NAs
expect_is(rt_model <- rtModel(rt_df = rt_df200,
time_matrix = time_matrix200,
price_diff = price_diff_l200,
estimator=structure('base', class='base')),
'rtmodel')
# Robust works but gives warning
expect_is(rt_model <- rtModel(rt_df = rt_df200,
time_matrix = time_matrix200,
price_diff = price_diff_l200,
estimator=structure('robust', class='robust')),
'rtmodel')
# Weighted works
expect_is(rt_model <- rtModel(rt_df = rt_df200,
time_matrix = time_matrix200,
price_diff = price_diff200,
estimator=structure('weighted', class='weighted')),
'rtmodel')
## Check severe sparteness
# Create data set
rt_df20 <- rt_df[1:20, ]
time_matrix20 <- rtTimeMatrix(rt_df20)
price_diff_l20 <- log(rt_df20$price_2) - log(rt_df20$price_1)
price_diff20 <- rt_df20$price_2 - rt_df20$price_1
# Base: Works, but gives message
expect_is(rt_model <- rtModel(rt_df = rt_df20,
time_matrix = time_matrix20,
price_diff = price_diff_l20,
estimator=structure('base', class='base')),
'rtmodel')
# Robust: Works, but gives warning from lmrob()
expect_warning(rt_model <- rtModel(rt_df = rt_df20,
time_matrix = time_matrix20,
price_diff = price_diff_l20,
estimator=structure('robust', class='robust')))
# Weighted: works but gives message
expect_is(rt_model <- rtModel(rt_df = rt_df20,
time_matrix = time_matrix20,
price_diff = price_diff20,
estimator=structure('weighted', class='weighted')),
'rtmodel')
})
## hpiModel.rtdata() after rtModel --------------------------------------------------------
context('hpiModel.rtdata(): after rtModel()')
test_that('hpiModel.rtdata works in both trim_model cases', {
expect_is(rt_model <- hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'base',
log_dep = TRUE,
trim_model=TRUE), 'hpimodel')
expect_is(rt_model <- hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'base',
log_dep = TRUE,
trim_model=FALSE), 'hpimodel')
expect_is(rt_model <- hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'base',
log_dep = FALSE,
trim_model=TRUE), 'hpimodel')
expect_is(rt_model <- hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'base',
log_dep = FALSE,
trim_model=FALSE), 'hpimodel')
expect_is(rt_model <- hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'weighted',
log_dep = TRUE,
trim_model=FALSE), 'hpimodel')
expect_is(rt_model <- hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'robust',
log_dep = FALSE,
trim_model=TRUE), 'hpimodel')
expect_true(is.null(hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'weighted',
log_dep = TRUE,
trim_model=TRUE)$model_obj$qr))
expect_true(!is.null(hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'weighted',
log_dep = TRUE,
trim_model=FALSE)$model_obj$qr))
})
test_that('hpiModel.rtdata outputs are correct', {
# Run a model of each estimator type
rt_model_base <- hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'base',
log_dep = TRUE,
trim_model=TRUE)
rt_model_robust <- hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'robust',
log_dep = TRUE,
trim_model=FALSE)
rt_model_wgt <- hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'weighted',
log_dep = FALSE,
trim_model=TRUE)
set.seed(123)
rt_model_wwgt <- hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'weighted',
log_dep = FALSE,
weights = runif(nrow(rt_df), 0, 1),
trim_model=TRUE)
# Estimatort
expect_is(rt_model_base$estimator, 'base')
expect_is(rt_model_robust$estimator, 'robust')
expect_is(rt_model_wgt$estimator, 'weighted')
# Coefficients
expect_is(rt_model_base$coefficients, 'data.frame')
expect_is(rt_model_robust$coefficients, 'data.frame')
expect_is(rt_model_wgt$coefficients, 'data.frame')
expect_true(nrow(rt_model_base$coefficients) == 84)
expect_true(max(rt_model_robust$coefficients$time) == 84)
expect_true(rt_model_wgt$coefficients$coefficient[1] == 0)
expect_false(identical(
rt_model_wgt$coefficients$coefficient,
rt_model_wwgt$coefficients$coefficient))
# Modelobj
expect_is(rt_model_base$model_obj, 'rtmodel')
expect_is(rt_model_robust$model_obj, 'rtmodel')
expect_is(rt_model_wgt$model_obj, 'rtmodel')
# Model spec
expect_true(is.null(rt_model_base$model_spec))
expect_true(is.null(rt_model_robust$model_spec))
expect_true(is.null(rt_model_wgt$model_spec))
# base price
expect_true(round(rt_model_base$base_price, 0) == 427785)
expect_true(round(rt_model_robust$base_price, 0) == 427785)
expect_true(round(rt_model_wgt$base_price, 0) == 427785)
# Periods
expect_is(rt_model_base$periods, 'data.frame')
expect_true(nrow(rt_model_base$periods) == 84)
expect_is(rt_model_robust$periods, 'data.frame')
expect_true(nrow(rt_model_robust$periods) == 84)
expect_is(rt_model_wgt$periods, 'data.frame')
expect_true(nrow(rt_model_wgt$periods) == 84)
# Approach
expect_true(rt_model_base$approach == 'rt')
expect_true(rt_model_robust$approach == 'rt')
expect_true(rt_model_wgt$approach == 'rt')
})
### modelToIndex --------------------------------------------------------------------
context('modelToIndex')
rt_model <- hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'base',
log_dep = TRUE,
trim_model=TRUE)
test_that('modelToIndex works', {
expect_is(modelToIndex(rt_model), 'hpiindex')
})
test_that('modelToIndex works with other estimatort and options', {
# Robust, LogDep=T, TrimModel=T
expect_is(modelToIndex(hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'robust',
log_dep = TRUE,
trim_model=TRUE)), 'hpiindex')
# Weighted, LogDep=T, TrimModel=T
expect_is(modelToIndex(hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'weighted',
log_dep = TRUE,
trim_model=TRUE)), 'hpiindex')
# Robust, LogDep=F, TrimModel=T
expect_is(modelToIndex(hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'robust',
log_dep = FALSE,
trim_model=TRUE)), 'hpiindex')
# Weighted, LogDep=T, TrimModel=F
expect_is(modelToIndex(hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'weighted',
log_dep = TRUE,
trim_model=FALSE)), 'hpiindex')
})
test_that('modelToIndex fails with an error',{
expect_error(modelToIndex(model_obj = 'abc'))
})
test_that('modelToIndex imputes properly, BASE model, LogDEP',{
model_base <- hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'base',
log_dep = TRUE,
trim_model=TRUE)
# Impute a beginning value
model_ex <- model_base
model_ex$coefficients$coefficient[2] <- NA_real_
expect_is(modelToIndex(model_ex), 'hpiindex')
expect_true(!is.na(modelToIndex(model_ex)$value[2]))
expect_true(modelToIndex(model_ex)$value[2] == 100)
expect_true(modelToIndex(model_ex)$imputed[2] == 1)
# Interpolate interior values
model_ex <- model_base
model_ex$coefficients$coefficient[3:5] <- NA_real_
expect_is(modelToIndex(model_ex), 'hpiindex')
expect_true(all(!is.na(modelToIndex(model_ex)$value[3:5])))
# Extrapolate end periods
model_ex <- model_base
model_ex$coefficients$coefficient[81:84] <- NA_real_
expect_is(modelToIndex(model_ex), 'hpiindex')
expect_true(all(!is.na(modelToIndex(model_ex)$value[81:84])))
expect_true(modelToIndex(model_ex)$value[80] ==
modelToIndex(model_ex)$value[84])
})
test_that('modelToIndex imputes properly, BASE model, LogDep=FALSE',{
model_base <- hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'base',
log_dep = FALSE,
trim_model=TRUE)
# Extrapolate a beginning value
model_ex <- model_base
model_ex$coefficients$coefficient[2] <- NA_real_
expect_is(modelToIndex(model_ex), 'hpiindex')
expect_true(!is.na(modelToIndex(model_ex)$value[2]))
expect_true(modelToIndex(model_ex)$value[2] == 100)
expect_true(modelToIndex(model_ex)$imputed[2] == 1)
# Impute interior values
model_ex <- model_base
model_ex$coefficients$coefficient[3:5] <- NA_real_
expect_is(modelToIndex(model_ex), 'hpiindex')
expect_true(all(!is.na(modelToIndex(model_ex)$value[3:5])))
# Extrapolate an end value
model_ex <- model_base
model_ex$coefficients$coefficient[81:84] <- NA_real_
expect_is(modelToIndex(model_ex), 'hpiindex')
expect_true(all(!is.na(modelToIndex(model_ex)$value[81:84])))
expect_true(modelToIndex(model_ex)$value[80] ==
modelToIndex(model_ex)$value[84])
})
test_that('modelToIndex imputes properly, Robust model, LogDEP = TRUE',{
model_base <- hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'robust',
log_dep = TRUE,
trim_model=TRUE)
model_ex <- model_base
model_ex$coefficients$coefficient[2] <- NA_real_
expect_is(modelToIndex(model_ex), 'hpiindex')
expect_true(!is.na(modelToIndex(model_ex)$value[2]))
expect_true(modelToIndex(model_ex)$value[2] == 100)
expect_true(modelToIndex(model_ex)$imputed[2] == 1)
model_ex <- model_base
model_ex$coefficients$coefficient[3:5] <- NA_real_
expect_is(modelToIndex(model_ex), 'hpiindex')
expect_true(all(!is.na(modelToIndex(model_ex)$value[3:5])))
model_ex <- model_base
model_ex$coefficients$coefficient[81:84] <- NA_real_
expect_is(modelToIndex(model_ex), 'hpiindex')
expect_true(all(!is.na(modelToIndex(model_ex)$value[81:84])))
expect_true(modelToIndex(model_ex)$value[80] ==
modelToIndex(model_ex)$value[84])
})
test_that('modelToIndex imputes properly, Weighted model, LogDep=FALSE',{
model_base <- hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'weighted',
log_dep = FALSE,
trim_model=TRUE)
model_ex <- model_base
model_ex$coefficients$coefficient[2] <- NA_real_
expect_is(modelToIndex(model_ex), 'hpiindex')
expect_true(!is.na(modelToIndex(model_ex)$value[2]))
expect_true(modelToIndex(model_ex)$value[2] == 100)
expect_true(modelToIndex(model_ex)$imputed[2] == 1)
model_ex <- model_base
model_ex$coefficients$coefficient[3:5] <- NA_real_
expect_is(modelToIndex(model_ex), 'hpiindex')
expect_true(all(!is.na(modelToIndex(model_ex)$value[3:5])))
model_ex <- model_base
model_ex$coefficients$coefficient[81:84] <- NA_real_
expect_is(modelToIndex(model_ex), 'hpiindex')
expect_true(all(!is.na(modelToIndex(model_ex)$value[81:84])))
expect_true(modelToIndex(model_ex)$value[80] ==
modelToIndex(model_ex)$value[84])
})
test_that('modelToIndex "max_period" cutoff works',{
model_base <- hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'weighted',
log_dep = FALSE,
trim_model=TRUE)
# Create shortened index
index_80 <- modelToIndex(model_base, max_period = 80)
expect_is(index_80, 'hpiindex')
expect_true(length(index_80$value) == 80)
})
### smoothIndex() -------------------------------------------------------------
context('smoothIndex()')
model_base <- hpiModel(model_type = 'rt',
hpi_df = rt_df,
estimator = 'base',
log_dep = TRUE,
trim_model=TRUE)
index_base <- modelToIndex(model_obj = model_base)
test_that('smoothing Function works with a variety of inputs',{
# HPI Index object
expect_is(index_smooth <- smoothIndex(index_obj = index_base,
order = 4),
'indexsmooth')
})
test_that('Errors are given when index is bad',{
# Non-sensical index
expect_error(index_smooth <- smoothIndex(index_obj = 'abc',
order = 3))
# Negative Order
expect_error(index_smooth <- smoothIndex(index_obj = index_base,
order = -3))
# Char Window
expect_error(index_smooth <- smoothIndex(index_obj = index_base,
order = 'x'))
# NA Window
expect_error(index_smooth <- smoothIndex(index_obj = index_base,
order = NA_integer_))
})
test_that('Returning in place works',{
# Add it to the Full HPI Object (to the hpiindex object)
expect_is(index_base <- smoothIndex(index = index_base,
order = 3,
in_place = TRUE),
'hpiindex')
expect_true('smooth' %in% names(index_base))
})
### rtIndex() wrapper ---------------------------------------------------------------
context('rtindex() wrapper')
test_that('Function works with proper inputs',{
# Full case
full_1 <- rtIndex(trans_df = sales,
date = 'sale_date',
price = 'sale_price',
trans_id = 'sale_id',
prop_id = 'pinx',
estimator = 'base',
log_dep = TRUE,
periodicity = 'monthly')
expect_is(full_1, 'hpi')
expect_true(full_1$model$estimator == 'base')
# Giving a 'trans_df' object
full_2 <- rtIndex(trans_df = sales_df,
price = 'sale_price',
trans_id = 'sale_id',
prop_id = 'pinx',
estimator = 'robust',
log_dep = TRUE)
expect_is(full_2, 'hpi')
expect_true(full_2$model$estimator == 'robust')
# Giving an 'rt_df' object
full_3 <- rtIndex(trans_df = rt_df,
estimator = 'weighted',
log_dep = TRUE)
expect_is(full_3, 'hpi')
expect_true(full_3$model$estimator == 'weighted')
})
test_that('Additional arguments in rtIndex() work',{
## rt Create arguments
# Min Date Model with Clip
mindate_index <- rtIndex(trans_df = sales,
date = 'sale_date',
price = 'sale_price',
trans_id = 'sale_id',
prop_id = 'pinx',
min_date = as.Date('2011-01-01'),
adj_type = 'clip')
expect_true(mindate_index$index$name[1] == '2011')
# Max Date Model with Adjust
maxdate_index <- rtIndex(trans_df = sales,
date = 'sale_date',
price = 'sale_price',
trans_id = 'sale_id',
prop_id = 'pinx',
max_date = as.Date('2015-12-31'))
expect_true(maxdate_index$index$name[7] == '2016')
# Periodicity
per_index <- rtIndex(trans_df = sales,
date = 'sale_date',
price = 'sale_price',
trans_id = 'sale_id',
prop_id = 'pinx',
periodicity = 'weekly')
expect_true(per_index$index$name[364] == 'week: 2016-12-11 to 2016-12-17')
# Sequence Only
seq_index <- rtIndex(trans_df = sales_df,
date = 'sale_date',
price = 'sale_price',
trans_id = 'sale_id',
prop_id = 'pinx',
seq_only = TRUE)
expect_true(nrow(seq_index$data) == 4823)
## HPI Model
# Trim Model
trim_index <- rtIndex(trans_df = rt_df,
trim_model=TRUE)
expect_true(is.null(trim_index$model$model_obj$qr))
# Log Dep & Robust
ld_index <- rtIndex(trans_df = rt_df,
estimator = 'robust',
log_dep = FALSE)
expect_true(ld_index$model$log_dep == FALSE)
expect_true(ld_index$model$estimator == 'robust')
## Model to Index
m2i_index <- rtIndex(trans_df = rt_df,
estimator = 'robust',
log_dep = FALSE,
max_period = 80)
expect_true(length(m2i_index$index$value) == 80)
})
test_that("Bad arguments generate Errors: Full Case",{
expect_error(rtIndex(trans_df = sales,
date = 'sale_price',
price = 'sale_price',
trans_id = 'sale_id',
prop_id = 'pinx',
estimator = 'base',
log_dep = TRUE,
periodicity = 'monthly'))
expect_error(rtIndex(trans_df = sales,
date = 'sale_date',
price = 'sale_price',
prop_id = 'pinx',
estimator = 'base',
log_dep = TRUE,
periodicity = 'monthly'))
expect_error(rtIndex(trans_df = sales,
date = 'sale_date',
price = 'sale_price',
trans_id = 'sale_id',
estimator = 'base',
log_dep = TRUE,
periodicity = 'monthly'))
expect_error(rtIndex(trans_df = sales,
date = 'sale_date',
trans_id = 'sale_id',
prop_id = 'pinx',
estimator = 'base',
log_dep = TRUE,
periodicity = 'monthly'))
expect_error(rtIndex(trans_df = sales,
date = 'sale_date',
price = 'sale_price',
trans_id = 'sale_id',
prop_id = 'pinx',
estimator = 'base',
log_dep = TRUE,
periodicity = 'xxx'))
expect_error(rtIndex(trans_df = sales[1, ],
date = 'sale_date',
price = 'sale_price',
trans_id = 'sale_id',
prop_id = 'pinx',
estimator = 'base',
log_dep = TRUE,
periodicity = 'monthly'), 'Converting transactions')
})
test_that("Bad arguments generate errort: trans_df Case",{
expect_error(rtIndex(trans_df = sales_df,
price = 'xx',
trans_id = 'sale_id',
prop_id = 'pinx',
estimator = 'base',
log_dep = TRUE))
expect_error(rtIndex(trans_df = sales_df,
price = 'sale_price',
trans_id = 'xx',
prop_id = 'pinx',
estimator = 'base',
log_dep = TRUE))
expect_error(rtIndex(trans_df = sales_df,
price = 'sale_price',
prop_id = 'pinx',
estimator = 'base',
log_dep = TRUE))
expect_error(rtIndex(trans_df = sales_df,
price = 'sale_price',
trans_id = 'sale_id',
prop_id = 'xx',
estimator = 'base',
log_dep = TRUE))
})
test_that("Bad arguments handling: rt_sales Case",{
# Bad estimatort default to 'base'
expect_true(rtIndex(trans_df = rt_df,
estimator = 'basex',
log_dep = TRUE)$model$estimator == 'base')
expect_error(rtIndex(trans_df = rt_df,
estimator = 'robust',
log_dep = 'a'))
expect_error(rtIndex(trans_df = rt_df,
estimator = 'robust',
trim_model = 'a'))
expect_error(rtIndex(trans_df = rt_df,
estimator = 'robust',
max_period = 'a'))
})
test_that("Smoothing in_place for 'hpi' object works",{
full_1 <- rtIndex(trans_df = sales,
date = 'sale_date',
price = 'sale_price',
trans_id = 'sale_id',
prop_id = 'pinx',
estimator = 'base',
log_dep = TRUE,
periodicity = 'monthly',
smooth = TRUE)
# In wrapper smoothing worked
expect_is(full_1$index$smooth, 'indexsmooth')
# Full HPI Object
expect_is(index_smooth <- smoothIndex(index_obj = full_1,
order = 6),
'indexsmooth')
# Add it to the Full HPI Object (to the hpiindex object) with new name
expect_is(full_1s <- smoothIndex(index = full_1,
order = 3,
in_place = TRUE),
'hpi')
expect_is(full_1s$index$smooth, 'ts')
expect_is(full_1s$index$smooth, 'indexsmooth')
})
#*****************************************************************************************
#*****************************************************************************************
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.