Hedonic Quality Adjustment Methods
For Clothes Dryers In the U.S. CPI
Paul R. Liegey
1Preface
As previously announced, BLS is extending the use of quality adjustments derived from
hedonic models in the CPI. A hedonic model decomposes the price of a consumer product into
implicit prices for each of its important features and components, thereby
providing an estimate of the value for each price-influencing feature and component.
Effective with the CPI for October 2000, BLS has extended hedonic quality adjustments
to washing machines and clothes dryers, two products in the Major Appliances item
stratum.2 The relative importance (share of weight), as of December 1999, for
this stratum was 0.205 percent in the CPI for all Urban Consumers (CPI-U) and 0.236
percent in the CPI for Urban Wage Earners and Clerical Workers (CPI- W). Within Major
Appliances, washing machines are estimated to represent 18 percent of the weight and
clothes dryers about 13 percent. Other products employing hedonic quality adjustments in
the Major Appliances item stratum are refrigerator/freezers and microwave ovens.3
The remaining products in this stratum—those that are not subject to hedonic quality
adjustment at this time—include freezers and stoves and ovens.
Of the three different approaches or methods that use the results from hedonic
regression models to quality adjust price indexes, the Bureau of Labor Statistics (BLS)
employs the 'matched model' method in its official indexes.4 This method
controls for quality changes based on the difference in product specifications or
characteristics between two items when a substitute observation, or quote, occurs in the
price index sample. It is important to note that under the 'matched model' approach only
substitution price changes are eligible for hedonic quality adjustments.
The U.S. Consumer Price Index (CPI) for Major Appliances would have remained
unchanged instead of the official index decline of 0.1 percent if hedonic quality
adjustment methods had been applied to clothes dryer (substitution) price changes from
October 1999 through June 2000 (see attachment 1). During the study period, clothes dryer
prices accounted for approximately 15 percent of the monthly Major Appliances
sample, and averaged only two (substitution) price changes per month (see attachment 2).
Background
The first clothes dryer was invented in France by Pochon in 1799 and was called ventilator
for drying clothes.5 This apparatus required wet clothes be wrung out by
hand, placed in a cylindrical metal drum pierced with holes or slits, and hand driven by a
crank over a fire. The clothes would dry out, but they might burn, and would always smell
strongly of smoke and possibly pick up soot from the smoke.
In 1930, J. Ross Moore built the first electronic drying device for clothes.6
Moore developed both gas and electric clothes dryers and patented his invention in 1936.
Financial difficulties led Moore to sell his clothes dryer design to the Hamilton
Manufacturing Company in 1937. Between 1938 and 1941, Hamilton Manufacturing Company sold
more than 6,000 clothes dryer units.
In 1947, the first post war year of full production, Hamilton Manufacturing Company,
General Electric and other entrants into the clothes dryer market sold approximately
41,000 and 20,000 units of electric and gas clothes dryers, respectively.7
Currently, factory shipments of clothes dryers total around six million units per year and
sales of electric units are three times as great as those of gas clothes dryers (see
attachment 3).
By the mid-1950s, about ten percent of U.S. households owned a clothes dryer at a price
of around $230 per unit—roughly $1600 in calendar year 2000 dollars. Ownership of clothes
dryers increased to about 45 percent of U.S. households by 1970 and the unit price at that
time was around $190—approximately $850 in calendar year 2000 dollars. By 1997, about 78
percent of U.S. households owned a clothes dryer with a unit price of around $340—about
$370 in calendar year 2000 dollars.8 Measuring in calendar year 2000 dollars,
the unit price for clothes dryers has declined by about 77 percent from the mid-1950s to
1997.
The steep decline in (modern) clothes dryer prices—as measured in calendar year 2000
dollars—reflects the increase in consumer purchasing power over these appliances while
numerous quality improvements have occurred over the last 50 years. Attachment 4 provides
a brief chronology of (modern) clothes dryer quality characteristic improvements since
their creation in the 1930s.
Clothes dryers were introduced into the CPI sample in 1963.9 Since their
inclusion in the CPI, clothes dryers have been represented in three item stratum price
indexes over four CPI revisions:
- With the release of the revised CPI in January 1964, clothes dryers became part of the Clothes
dryers, electric item stratum. The index level for this item stratum in December 1963
was 100.0 (its base year), and increased to 104.0 by December 1970 — an overall price
index increase of 4.0 percent for the six year period. In January 1971, the Clothes
dryers, electric item stratum was calculated on a new base year (1967=100). From
December 1970 through December 1977, the index for Clothes dryers, electric climbed
from 110.1 to 155.1 — an overall price index increase of 40.9 percent for the seven year
period.
- With the release of the revised CPI in January 1978, clothes dryers became part of the Laundry
equipment item stratum. The index level for this item stratum in December 1977 was
100.0 (its base year), and increased to 147.8 by December 1986 — an overall price index
increase of 47.8 percent for the nine year period.
- With the release of the revised CPI in January 1987, clothes dryers remained part of the
Laundry equipment item stratum. The index level for this item stratum in December 1986
was 104.0 (with a base period of 1982-84=100), and increased to 113.4 by December 1997 —
an overall price index increase of 9.0 percent for the 11 year period.
Finally, with the release of the revised CPI in January 1998, clothes dryers became part
of the Major Appliances item stratum. The index level for this item stratum in
December 1997 was 100.0 (its base year), and through October 2000 is 96.5 — an overall
price index decline of 3.5 percent for the most recent 34 month period.
Clothes dryers were selected as a product that would benefit from hedonic regression
modeling since manufacturers provide a selection of types, brands, sizes, and features. In
the August 2000 Consumer Report issue, the latest annual review on clothes dryers
notes that "compared with washers, dryers are relatively simple. Their major
distinctions are how they heat the air (gas or electric) and how they're programmed to
shut off once the load is dry (thermostat or moisture sensor). Both will affect how much
you'll pay to buy and run your machine."10
Newer models of clothes dryers are included for price index calculations only if older
models being used in the price index sample are no longer available (in the CPI outlets)
for CPI pricing. When a newer model replaces an older model in the CPI, the price change
that is used in the index is referred to as substitution price change. Substitution price
change can be either "pure" (directly compared or quality adjusted) or
"imputed" (not compared).
From October 1999 through June 2000 Major Appliances employed, on average, 218
price changes to calculate the U.S. level monthly price index (see attachment 2).
Approximately nine percent, or 19, of these price changes were substitution price changes.
In comparison, clothes dryers accounted for, on average, 32 of the 218 price changes used
to calculate the Major Appliances index, and averaged (just) two substitutions per
month from October 1999 through June 2000.
Data
Sample selection for the Hedonic Model
The official CPI sample of clothes dryer prices used to calculate the Major
Appliances CPI was too small for hedonic regression estimation. Using a process that
mimics the official CPI sample selection process, an additional sample of 194 consumer
businesses, or outlets, was chosen to augment the official CPI sample for clothes dryers.
This additional outlet sample was used to select a sample of clothes dryer prices. The
supplementary sample was used only for estimating the hedonic regression model for clothes
dryers.11
CPI field economists were instructed to collect a total of 400 clothes dryer prices in
the sample of 194 additional outlets. Individual clothes dryer brand and models were
selected by grouping all clothes dryers in a particular outlet into two groups—the
"standard" clothes dryers and "better model" or higher quality clothes
dryers. Once categorized into these two groups, the CPI field economist was instructed to
select a "good selling" clothes dryer from each of the "standard" and
"better model" groups.
About 69 percent of the additional sample price quotes that were collected for clothes
dryers had price and characteristic data that could be used in the regression model. The
most common reason that CPI field economists could not collect the additional sample price
quotes for clothes dryers was respondent refusal.12 A total of 341 prices—and
characteristic descriptions—were used to estimate the hedonic model for clothes dryers.
This total sample consists of 64 official CPI observations and 277 additional sample
observations.
Price and Characteristics Data for the Hedonic Model
All of the price and characteristics data used for the clothes dryer hedonic model were
captured on CPI data collection documents, or checklists, for this item (see last
attachment).
The prices that were collected for the clothes dryer sample represent "retail
offer" prices. As the name suggests, a retail offer price represents what a consumer
business is willing to sell an item for which may, or may not, differ from the transaction
price—what a consumer actually paid for the item. Retail offer prices, like transaction
prices, may change through time depending on whether the item being sold is offered at a
"regular" price or a "sale" price.
The set of quality attributes collected for each of the 341 clothes dryers in the
sample are represented on the CPI checklist for this item (see last attachment). In each
of the quality characteristic categories, CPI field economists selected the specific
characteristic element that best described the item they were pricing. For example if an
electric clothes dryer with a (drum volume) capacity of 6.0 cubic feet had been selected
by the field economist, this would be designated on the CPI data collection document by
selecting the A1 and D5 specification elements—see last attachment.
When possible, secondary source information such as manufacturer websites and consumer
information magazines—including Consumer Digests and Consumer Reports—were
used to verify the accuracy of the characteristic data collected on the CPI checklist for
clothes dryers.
Model
The hedonic model that is specified for clothes dryers in this study uses the quality
characteristics that were collected on the CPI checklist for this item (see last
attachment). The independent variables in the model fall into three groups similar to the
hedonic model for dishwashers referenced by Greenlees (2000).13 In particular,
technical appliance characteristics, product brands and outlet types—as well as other
control type variables—are tested in iterative regressions to determine their impact on
the natural logarithm of price, the dependent variable. Twelve preliminary regressions are
presented in Attachment 5 to give the reader a sense of the model development through
completion.
The CPI prices that are collected in this sample represent "retail offer"
prices, and approximately 38 percent of these prices were collected "on sale."
The mean price for all clothes dryers in the sample is $488.78. The mean price for
"regular" priced clothes dryers in the sample is $511.36 and the mean price for
"sale" priced clothes dryers is $452.86. Since type of price (that is, regular
or sale) is thought to have an impact on the overall price level, a dummy variable for
sale price is included in the model to capture this effect, and its expected coefficient
sign is negative.
A priori
expectations about which clothes dryer characteristics influence price
were developed, when possible, from industry information, manufacturer websites, and
consumer information magazines and websites.
Clothes Dryer Types
Two types of clothes dryers—electric and natural gas—are sold in
today's market. Electric models are sold more frequently by retailers and account
for approximately 65 percent of the clothes dryers in the sample. Typically, retail offer
prices for electric dryers are less then gas dryers. An August 2000 Consumer
Reports study on clothes dryers finds that "the saving in energy costs should
cover the slightly higher purchase price (of gas models) within the first year."14
Mean prices for the 223 electric models and 118 gas models in the hedonic
sample were $418.54 and $620.42, respectively. By way of comparison, the Consumer
Reports article (referenced above) on clothes dryers shows that the electric
models they tested retailed for an average of $489.09 while the natural gas models
retailed for an average of $560.00. Dummy or indicator variables are created for both
types of clothes dryers. The gas variable included in preliminary hedonic models is
expected to have a large, positive impact on price.
Clothes Dryer Capacity
A fact sheet issued by the University of Nebraska maintains that "differences
between clothes dryers are largely capacity, price and dryer size (physical dimensions of
the dryer)."15 Manufacturers and retailers advertise drum capacity
in terms of cubic feet. Drum capacity can assume a wide range of numeric
values—see the CPI checklist for clothes dryers (last attachment). To accommodate this
potential wide range of values, a continuous variable is created for capacity (in
cubic feet). It is assumed that the price of clothes dryers increases with increasing
values of this variable. For example, a 7.0 cubic foot dryer is assumed to sell for a
higher price than a 5.0 cubic foot dryer ceteris paribus.
Clothes Dryer Dry Cycles
As noted in Shepler (2001) "the cycle category was difficult for the data
collectors since manufacturers do not always list every single (wash) cycle and they also
use different terminology for similar cycles."16 The dry cycle
specification category on the clothes dryer checklist—see last attachment—proved to be
just as problematic as the wash cycle category for clothes washers in terms of obtaining a
complete and consistent description across specific brands and model numbers. In addition
to the eight dry cycles listed on the checklist, there is also a specification category
for number of dry cycles. Manufacturers and retailers tend to do a better job at
reporting the number of dry cycles rather than listing each cycle. Secondary source
information was used to verify / overwrite the information in the sample data set and a
variable for number of dry cycles is created. It is assumed that dryers with
a greater number of dry cycles provide consumers with greater drying functionality
and have a positive impact on price.
Preliminary models are specified with two dummy variables, sale price and gas
dryer, and two continuous variables, capacity (in cubic feet) and number of
dry cycles (see Attachment 5, Iterative regressions 1 and 2). The first model
specification with sale price, gas dryer and capacity (in cubic feet)
proved to explain a significant portion of the variation in (the natural log of) price
with an R2 of slightly less than 54 percent. The magnitude, direction and
significance of the parameter estimates in this first preliminary model generally
conformed to a priori expectations
Inclusion of the number of dry cycles variable resulted in somewhat lower
tolerance values for the capacity (in cubic feet) and number of dry cycles
parameter estimates and indicates that multicollinearity might be present in the model.
Further investigation revealed that the pearson correlation coefficient, or measure of
collinearity, for these two variables is positive and strong at 0.74—they tend to move
together and can serve as a proxy for each other. The existence of multicollinearity
causes the standard errors of the correlated variables to increase and the associated
parameter estimates to be imprecise.17 Subsequent variations of these
preliminary models—not included in Attachment 5—revealed that the variable for capacity
(in cubic feet) provides a better overall fit for the model; therefore, this variable was
included in the final model.
Clothes Dryer Drying Mechanisms, Number of Temperature Settings and Control Types
Other 'technical' clothes dryer characteristics are included in the "Drying
Mechanisms," "Number of Temperature Settings," and "Control
Types" specification categories on the CPI checklist (see last attachment) and are
next considered in specifying the hedonic regression model for clothes dryers.
According to the Department of Energy, "the best dryers have moisture sensors
in the drum for sensing dryness, while most only infer dryness by sensing the temperature
of the exhaust air. Compared with timed drying, you can save about 10% with a temperature
sensing control, and 15% with a moisture sensing control."18 The moisture
sensor characteristic is found in 54 percent of the clothes dryer sample while the temperature
sensor and timed drying characteristics occur throughout the entire sample. It
is assumed that the presence of a moisture sensor would have a positive impact on
price.
The number of temperature settings specification category on the CPI
checklist—see last attachment—designates how many drying temperatures a user can select
from on a specific brand and model.
Most dryers have at least three temperature settings: regular, low, and cool. The more
sophisticated the dryer, the greater the choice one has in temperature setting.
Top-of-the-line models may have infinitely variable temperature settings. This allows
users to fine-tune the temperature to the type of fabric being dried. Clothes dryers with
three to five temperature settings comprise more than 80 percent of the sample.
Secondary source information was used to verify / overwrite the information in the sample
data set and a continuous variable for number of temperature settings is created.
It is assumed that dryers with a greater number of temperature settings provide
consumers with greater drying functionality and have a positive impact on price.
In the "Control Types" specification category, dummy variables were created
for electronic controls and push button or rotary dial manual controls.
While most of the clothes dryers in today's appliance market possess push button or
rotary dial manual controls some manufacturers are offering models with touch
sensitive or electronic controls on their top of the line dryers. Electronic
controls offer conveniences such as stored, customized temperature / cycle settings
and easy to read displays. More than 90 percent of the clothes dryer price sample
contained models with push button or rotary dial manual controls while the
remaining 10 percent reflected electronic controls. It is expected that the electronic
controls variable has a positive impact on price.
The moisture sensor, number of temperature settings and electronic
controls variables are tested in iterative regressions 3, 4 and 5—see Attachment 5.
All three variables provide a positive and significant impact on (the natural log of)
price while increasing the explanatory power of the successive hedonic models. The pearson
correlation coefficient, or measure of collinearity, for the moisture sensor and number
of temperature settings variables is positive and somewhat high at 0.55, but is deemed
acceptable and the level of multicollinearity does not warrant exclusion of either
variable. Representation of the "Drying Mechanisms," "Number of Temperature
Settings," and "Control Types" specification categories from the CPI
checklist (see last attachment), increases the overall explanatory power of the
preliminary hedonic models from an R2 of slightly less than 54
percent—iterative regression 1—to an R2 of slightly more than 74
percent—iterative regression 5.
Clothes Dryer Brands
Brand
is the next category of quality characteristics considered for use to specify
the clothes dryer hedonic model. Of the 17 brands listed on the CPI checklist for this
item (see last attachment), dummy or indicator variables are created for 11 brands that
are collected in the clothes dryer hedonic sample. Triplett and McDonald (1977) note that
"the interpretation and treatment of company (
brand) effects in hedonic
regressions is a perplexing problem for which there is not a straightforward
solution."
19 In this study,
brand is thought to serve as a proxy
for subtle and / or difficult-to-collect quality characteristics that represent the
general level of quality for a given
brand. Triplett and McDonald (1977) go on to
include
brand in their refrigerator hedonic regressions "because statistical
tests revealed them relevant and important."
20 Other hedonic studies on
appliances have also found it beneficial to include
brand.
21Some a priori information about clothes dryer brands is found at
ApplianceAdvisor.com, a consumer information website, which references 19 different brands
of dryers.22 Of the 11 brands that are collected in the clothes dryer
hedonic sample, 10 are included in iterative regression 6—see Attachment 5—excluding the
most frequently occurring brand which accounted for a little more than 44 percent
of the sample. In addition, three of the 10 brands included in preliminary models
are excluded since they exhibited statistically insignificant parameter estimates.
Inclusion of brand variables in preliminary model specifications yields a better
explanatory fit with an R2 of slightly more than 84 percent—iterative
regressions 6 through 12—from an R2 of slightly more than 74
percent—iterative regression 5. The seven brands that remained in iterative
regressions 7 through 12 emerge as two distinct groups—the "medium" and
"low" quality groups.23 These two distinct groups are clearly
delineated in iterative regression 12 where the parameter estimates for each of the seven
individual brands are listed in descending magnitude. The "medium" and
"low" quality groups are compared to the "base" group—the
"high" quality brands that are the most frequently occurring and not
included in the preliminary models.
An F test for brand equivalence is conducted for the brands in both the
"medium" and "low" quality groups—see Attachment 5 after iterative
regression 12. For both groups, the F tests fail to reject the null hypothesis that the brands
are equivalent. The parameter estimates for both the "medium" and
"low" quality groups are (highly) statistically significant and add just as much
explanatory power to the model as the individual brands variables. Both the "medium
quality" and "low quality" brand groups were included in the
final model specification.
Other Clothes Dryer Physical Characteristics and Features
Both the "Number of Motor Speeds" and "Drying Drum Material"
specification categories on the CPI checklist (see last attachment) proved to be difficult
to collect in the field. The data for these specification categories had little variation
when reported and is suspect since manufacturer websites did not consistently report this
information in their product descriptions for clothes dryers. Subsequently, a decision was
made not to use the data from these specification categories.
Unlike washing machines, all clothes dryers are front loading.24 The
"Door Style" category is included on the CPI checklist to collect information
about whether the specific clothes dryer being priced is a side-pull (opens left or
right) model or a front-pull (opens down) model. Side-pull doors can usually
be purchased as either a left-swing or a right-swing door. A front-pull or hamper
door saves lateral space and offers an additional 'shelf' on which to lay clothes for
cooling down or folding. Dummy variables are created for both the side-pull and front-pull
variables. The front-pull variable is assumed to have a positive impact on price
and is tested in iterative regression 7. While the front-pull parameter estimate is
positive and statistically significant, it has a (somewhat) high correlation with capacity
(in cubic feet)—the pearson correlation coefficient is positive with magnitude of
0.64—and does not add any substantial explanatory power to the model. A judgment was made
to exclude the "Door Style" specification category from the model.
Manufacturers and retailers of clothes dryers offer and advertise a variety of features
on their products. Clothes dryer features are represented in the "Features"
specification categories on the CPI checklist (see last attachment). Dummy or indicator
variables are created for all characteristics in this specification category. Each of the
dummy variables is tested to see whether the overall fit of the model could be enhanced.
Basic preliminary models used to test each of the features are presented in Attachment 5,
iterative regressions 8 through 11.
Of the seven dummy variables that are created for the "Features"
specification category, all are excluded from the final regression model because of their
poor performance in preliminary and subsequent regression models. In particular, the lint
filter indicator and sound insulation variables are excluded from the final
model because of statistical insignificance. Sound insulation is another example of
a characteristic that is difficult to collect due to varying terminology and limited
product descriptions in retail outlets. Secondary source information was used to verify /
overwrite the sound insulation information collected for the sample data set but
the parameter estimate for sound insulation remain insignificant.
The drying rack, drum light and end of cycle signal variables are
excluded from the final model because of multicollinearity with the capacity (in
cubic feet) and moisture sensor variables. Inclusion of the drying rack, drum
light and end of cycle signal variables produce (significant) movement in the
magnitude of the parameter estimates for the capacity (in cubic feet) and moisture
sensor variables—see iterative regression 9. The capacity (in cubic feet) and moisture
sensor variables are thought to be more germane to the final model specification and
the drying rack, drum light and end of cycle signal variables are
excluded.
The easy access lint filter variable is also excluded from the final model
because it was assumed a priori that this variable should have a positive impact on
price but its parameter estimate is consistently negative—see iterative regressions 8
through 11. Finally, the leveling legs feature was collected for 55 percent of the
sample but increased to more than 90 percent of the sample when secondary source
information was used to verify / overwrite this quality characteristic. Although the
parameter estimate for leveling legs remained positive and statistically
significant it was thought that this featured occurred even more frequently than reflected
in the sample and is excluded from the final model.
Clothes dryer features—like those collected on the CPI checklist for this product—are
frequently touted by retailers and manufacturers as value enhancing but have not
contributed (substantially) to the price, or value, of clothes dryers in this study.
Clothes Dryer Miscellaneous Specifications
Other CPI checklist specification categories that are examined for hedonic model
development include manufacturer warranty, country of origin, as well as clothes
dryer color. These CPI categories of specifications are found to have little
variation in the clothes dryer sample. In particular, almost all clothes dryers that are
sold in today's market come with a standard one year parts and labor manufacturer
warranty. About 10 percent of the clothes dryer sample reflected a warranty other than one
year parts and labor. Dummy variables were created for these other types of warranties
but each of the parameter estimates for these other warranties tested statistically
insignificant—see iterative regression 10. Subsequently, manufacturer warranty is
excluded from the final hedonic model for clothes dryers.
Similarly, the country of origin specification category on the CPI checklist is
tested to see if the hedonic model for clothes dryers could be fitted with any of these
variables. Country of origin represents the country in which the product—in this
case clothes dryers—was constructed and is believed to serve as a proxy for the quality
of a good and service. In the sample used in this study, over 98 percent of the 341
clothes dryers were constructed in the USA. The remaining clothes dryer models in
the sample were made in Canada.
Finally, about 90 percent of the clothes dryers in the sample are white while
about three percent are almond and the remaining seven percent are described as graphite.
It was thought that since some outlets charge extra for colors other than white,
that color might be a price factor.
Like the manufacturer warranty variables, both the country of origin and color
variables do not seem to have a statistically significant impact on price—see iterative
regression 11. Both the country of origin and color characteristics are
excluded from the final hedonic model.
On Clothes Dryer Energy Efficiency
As the reader may have already noticed, clothes dry energy efficiency does not appear
(explicitly) on the CPI checklist nor in the hedonic model for this product. As noted by
the Department of Energy:
The efficiency of a clothes dryer is measured by a term called the energy factor. It is
somewhat similar to the miles per gallon for a car, but in this case the measure is pounds
of clothing per kilowatt-hour of electricity. The minimum rating for a standard capacity
electric dryer is 3.01. For gas dryers the minimum energy factor is 2.67. The rating for
gas dryers is provided in kilowatt-hours though the primary source of fuel is natural gas.
Unlike most other types of appliances, energy consumption does not vary significantly
among comparable models of clothes dryers. Clothes dryers are not required to display
EnergyGuide labels.25
Proxy measures for clothes dryer efficiency include clothes dryer type—electric
or gas—and clothes dryer drying mechanisms—moisture sensors versus temperature
sensors and timed drying.
Control Variables
Various control variables are tested representing region of the country, city
size, and type of business as defined in the CPI. The most consistently
performing and statistically significant control variables that are included in the final
clothes dryer model are hardware outlets and warehouse outlets. The purpose
of control variables is to minimize the unexplained variation that might remain
after the model has been fitted with price determining characteristics.
Attachment 5, iterative regressions 1 through 12, are included in this study to give
the reader a sense of how the hedonic model for clothes dryers progressed as more
categories of CPI quality characteristics are considered for model inclusion. Iterative
regressions—both included and not included in Attachment 5—are performed until the
remaining parameter estimates in the model below exhibited relative robustness to the
inclusion and deletion of other variables not included. The direction and magnitude of the
parameter estimates seem reasonable, and the statistics pertaining to fit, significance,
and collinearity are within generally accepted limits.
Variable Category
|
Variable Name |
Parameter Estimate |
T Statistic |
Tolerance |
|
Intercept |
5.295815 |
46.235 |
|
Type of Price: |
Sale Price |
-0.103660 |
-7.257 |
0.8797381 |
Dryer Type: |
Gas |
0.170917 |
10.641 |
0.7268179 |
Electric |
Base |
|
|
Drum Capacity (in cubic feet) |
Capacity |
0.113937 |
6.943 |
0.5924369 |
Drying Mechanism |
Moisture sensor |
0.082960 |
3.920 |
0.3817178 |
Temp sensor / Timed dry |
Base |
|
|
Number of Temperature Settings |
Number of temperatures |
0.025726 |
6.255 |
0.5668198 |
Dryer Control Types |
Electronic controls |
0.181387 |
7.544 |
0.8441393 |
Rotary / Push button controls |
Base |
|
|
Brand: |
High quality |
Base |
|
|
Medium quality |
-0.200536 |
-10.915 |
0.6747474 |
Low quality |
-0.305682 |
-12.153 |
0.6655662 |
Type of Outlet: |
Hardware outlet |
-0.110854 |
-3.038 |
0.9439328 |
Warehouse outlet |
-0.292446 |
-3.366 |
0.9692976 |
R2=0.8481 |
Adjusted R2=0.8436 |
F value=184.87 |
Number of observations=341 |
Hedonic Quality Adjustments and Price Index Simulation
BLS employs the 'matched model' method of quality adjustment in its official indexes.
This method controls for quality changes based on the difference in product specifications
or characteristics between two items when a substitute observation, or quote, occurs in
the price index sample. It is important to note that under the 'matched model' approach only
substitution price changes, or quotes, are eligible for hedonic quality adjustments.
During the study period from October 1999 through June 2000, a total of 20 clothes
dryer substitution price changes—an average of just two per month—occurred in the Major
Appliances price index sample. For each clothes dryer substitution price quote,
differences in the specification or characteristic data of the old and new items were
identified to see if the parameter estimates in the hedonic model could be utilized to
quality adjust the official price change.
Fifty percent, or 10, of the 20 clothes dryer substitute price changes in the study are
quality adjusted based on specification differences between substitute items. The most
common type of quality adjustment performed for clothes dryers in this study is for
changes in capacity—represented in 80 percent of the quality adjusted
(substitution) prices. Other quality differences such as changes in number of
temperature settings and brand specifications are also adjusted for in the
hedonic indexes. The table below provides a summary of mean price changes for clothes
dryer substitutions in both the official and quality adjusted Major Appliances
indexes.
Summary of mean price changes for Clothes Dryer substitutions
Type of Substitution |
Published Index |
|
Quality Adjusted Index |
|
Number |
Mean Price Change |
Number |
Mean Price Change |
All Substitutions |
20 |
+10.41 % |
20 |
+9.29 % |
Comparable |
19 |
+10.75 % |
10 |
+12.59 % |
Quality Adjusted |
0 |
NA |
10 |
+5.98 % |
Non-comparable (Imputed) |
1 |
+3.90 % |
0 |
NA |
Substitution price change can be either "pure" (directly compared or quality
adjusted) or "imputed" (non-comparable). The empirical results above reveal that
not only was all "imputed" price change replaced with "pure" price
change in the quality adjusted index but that almost half of the "comparable"
substitution price changes in the published index were made "more comparable or
pure" in the quality adjusted index. In addition, the mean price change for all
substitutions is lower after quality adjustment than in the published indexes. This result
implies that some price increases due to quality change may have been reflected in the
published indexes.
Monthly price indexes are simulated for October 1999 through June 2000 to determine the
impact of quality adjusted clothes dryer substitution price changes on the Major
Appliances CPI. First the published indexes, or without quality adjustment indexes,
are recreated by Statistical Analysis System (SAS) computer programs using historical CPI
data. The duplication of the published indexes provides a "control" environment
from which the quality adjusted Major Appliances indexes are compared.
U.S. level price indexes, such as those examined in this study, are obtained by summing
elementary, or local area, price indexes using aggregation weights derived from the
Consumer Expenditure Survey (CE). The results of the index simulation, with and without
hedonic quality adjustments, are presented in Attachment 6. In addition, graphical
representations of the simulated price indexes for Major Appliances are presented
in Attachments 1 and 7.
Attachments 1 and 6 indicate that the U.S. CPI for Major Appliances would have
remained unchanged instead of the official index decline of 0.1 percent if hedonic quality
adjustment methods had been applied to clothes dryer (substitution) price changes from
October 1999 through June 2000. Given the small representation that clothes dryer
substitution price changes have in the Major Appliances CPI during the study
period—an average of just two price quotes per month—the impact at the U.S. level is
relatively large.
Attachments 6 and 7 illustrates the differences between the published and quality
adjusted indexes on a monthly basis for the nine-month period in the study—October 1999
through June 2000. Comparisons of the quality adjusted and published Major Appliances
price indexes reveal the following: In months when the index increased, the quality
adjusted index rose at the same rate for three months (November, March and April). In
addition, in the remaining six months when the index fell, there are three months
(February, May and June) in which both indexes fell at the same rate; two months (December
and January) in which the quality adjusted index fell (slightly) less than the published
index; and one month (October) in which the quality adjusted index fell (slightly) more
than the published index.
Two factors may account for the empirical results reported on in this study:
First, the small number of clothes dryer substitution price quotes that are quality
adjusted may have been too few to make a representative impact on the Major Appliances
indexes in 1999-2000. Clothes dryer substitution price quotes accounted for less than one
percent—on average, two of 218 prices—of the monthly CPI sample for Major Appliances
from October 1999 to June 2000 and only half of these dryer substitutions had their price
changes adjusted using the hedonic technique.
Second, CPI data collection procedures direct BLS field economists to select
substitution or replacement items that are the "same or similar" in quality as
the old item they had been pricing. This procedure tends to yield substitution price
quotes that have fewer bona fide characteristic changes (between substitute items)
than might have occurred if the procedures had instructed field economists to collect
(substitute item) data for the most technologically advanced or best selling clothes
dryers.
This second factor is important in developing expectations as to the direction and
magnitude of quality adjusted indexes when compared to indexes without quality adjustments
for consumer appliance goods. If BLS field economists are instructed to substitute to the
best selling or most technologically advanced clothes dryer products, one might expect
that the Major Appliances indexes with and without hedonic quality adjustments
would diverge further from each other than in this study.
BLS is considering additional ways to more quickly bring a greater number of new goods
quotes into the CPI rather than just relying on the current Telephone Point of Purchase
Survey (TPOPS) rotations. Lane (2000) provides a summary of additional methods for
bringing new goods into CPI samples more quickly.26 In particular, both the directed
item rotation and directed item replacement methods of updating price index
samples instruct field economists to "select a new set of (sample) items representing
a more recent period's purchases" for target groups of goods or services that are
constantly changing in quality with successive generations of product introductions.
Conclusion
The clothes dryer hedonic model developed in this study represents a snapshot of how
the average consumer values quality for clothes dryers in todays appliance market.
Iterative regressions for clothes dryers are included in this study to give the reader a
sense of hedonic model development as more categories of CPI characteristics were
considered for model inclusion. The current rate of quality change for this consumer
appliance product is not as fast as other retail products—for example, cellular and
wireless telephones or DVD players in the consumer electronics market.
The parameter estimates in the final hedonic model for clothes dryers exhibit relative
robustness to the inclusion and exclusion of other variables not included in the final
model. The direction and magnitude of the parameter estimates seem reasonable, and the
statistics pertaining to fit, significance and collinearity are within expected limits. It
is being used in conjunction with commodity analyst judgment to quality adjust CPI quote
level clothes dryer substitution data when possible.
Notes
- Thanks to Charles Fortuna, Mary Kokoski and Rick Devens for helpful comments.
- See BLS October 2000 News Release for the Consumer Price Index under "Extending the
use of hedonic models to adjust prices for changes in quality" heading on page 4;
ftp://ftp.bls.gov/pub/news.release/History/cpi.11162000.news.
- See BLS July 2000 News Release for the Consumer Price Index under "Extending the
use of hedonic models to adjust prices for changes in quality" heading on page 4;
ftp://ftp.bls.gov/pub/news.release/History/cpi.08162000.news. Also, papers describing
hedonic quality adjustments in the US CPI for refrigerator/freezers and microwave ovens
can be found on the BLS website at http://www.bls.gov/cpirfr.htm and
http://www.bls.gov/cpimwo.htm, respectively.
- Silver (1998) notes that there are " three different approaches to the use of
hedonic regressions for measuring quality-adjusted price changes. The first
complements the existing matched models approach generally used by statistical offices, by
helping to identify key quality characteristics and, when matches are not available,
providing adjustment factors to allow like to be compared with
like. The second is the direct method, found in the academic
literature, which uses the coefficients on the dummy variables for time in a hedonic
regression as estimates of quality-adjusted price changes. The third method
requires quite extensive data for the compilation of exact hedonic price
indices as defined from economic theory."—Page 1 from reference below.
- Source, The World Almanac Book of Inventions, distributed in the United States by
Ballantine Books (ISBN 0-911818-96-0), a division of Random House, Inc. Copyright 1985,
under the Clothes dryer heading on page 167.
- Source, "Appliance Histories Milestones in Appliance History: The White Goods
Story," Association of Home Appliance Manufacturers (AHAM); December 1998; page 2;
http://www.aham.org/cf-dbm/business/businessfirst.cfm (accessed June 2000).
- The author would like to thank Jill A. Notini, Communications Manager, of the
Association of Home Appliance Manufacturers (AHAM) for background materials on the
development of clothes dryers including data on factory shipments (including exports) from
1946 to present. Clothes dryer background materials and data on factory shipments
(including exports) from 1946 to present are available from the author upon request.
- The information about per unit clothes dryer prices and percent of ownership in US
households of clothes dryers are from Exhibits 5 and 8, respectively, in the 1997 Dallas
Federal Reserve Annual Report entitled, "Time Well Spent: The Declining Real Cost of
Living in America," by W. Michael Cox and Richard Alm found at
http://www.dallasfed.org/htm/pubs/pdfs/anreport/arpt97.pdf (accessed June 1998). The
conversion to calendar year 2000 dollars was done using the BLS inflation calculator found
at http://data.bls.gov/cgi-bin/cpicalc.pl.
- See Gordon, The Measurement of Durable Goods Prices; 1989, page 294.
- See Consumer Reports August 2000; "Clean Machines: Washers and Dryers,"
page 36.
- For additional information about the hedonics project on quality change in the U.S. CPI
see Fixler, Fortuna, Greenlees and Lane (1999), "The Use of Hedonic Regressions to
Handle Quality Change: The Experience in the U.S. CPI," Presented at the Fifth
Meeting of the International Working Group on Price Indices; August 1999; Pages 1-20.
http://www.statice.is/ottawa/bls.rtf (accessed November 1999).
- The collection of hedonic price data by CPI field economists coincided with the
collection of official CPI price data and TPOPS price data. Reports from the field
indicated that some respondents simply did not have the time to assist CPI field
economists with hedonic data collection.
- See John Greenlees, "Consumer price indexes: methods for quality and variety
change," in Statistical Journal of the United Nations ECE 17 (2000), pages 59-74.
Reference to hedonics on page 64.
- With recent spikes in U.S. natural gas prices in the winter of 2000-2001, the long term
operational savings of using a natural gas dryer maybe be diminished or eliminated.
- Source: "Making Decisions: Buying a Clothes Dryer," Issued March 1998 by the
University of Nebraska in cooperation with the U.S. Department of Agriculture; page 2;
http://www.ianr.unl.edu/pubs/housing/nf348.htm (accessed December 1999).
- Source: Shepler (2001) in "Development and Application of a Hedonic Regression
Model For Washing Machines In the U.S. CPI," in the Data and Regression Model
section; at http://www.bls.gov/cpiwas.htm
- Source: Shepler (2000) in "Developing a Hedonic Regression Model For Refrigerators
in the U.S. CPI," in the Data and Regression Model section on page 3;
http://www.bls.gov/cpirfr.htm.
- Source: "Tips for Buying a New Clothes Dryer," Department of Energy's Office
of Codes and Standards Energy Efficient Appliances; Updated May 24, 2000;
http://www.eren.doe.gov/buildings/consumer_information/dryers/drytips.html (accessed June
2000).
- Triplett, Jack E., and Richard J. McDonald. 1977. "Assessing the quality error in
output measures: The case of refrigerators," Journal of Review of Income and Wealth,
No. 23 (June), pages 127-56.
- Ibid; page 145.
- In addition to the dishwasher study referenced in Greenlees (2000)—see 1) Bascher and
Lacroix, "Dishwashers and PCs in the French CPI: hedonic modeling, from design to
practice," August 1999; http://www.statice.is/ottawa/lacroix.rtf. See also, 2)
Hoffmann, "The Treatment of Quality Changes in the German Consumer Price Index,"
August 1999; http://www.statice.is/ottawa/hoffmann.rtf. See also, 3) Silver and Heravi,
"The Measurement of Quality-Adjusted Price Changes," September 2000;
http://www.nber.org/~confer/2000/criwf00/silver.pdf. See also, 4) Shepler,
"Developing a Hedonic Model For Refrigerators in the U.S. CPI," August 2000;
http://www.bls.gov/cpirfr.htm. See also, 5) Liegey, "Hedonic Quality Adjustment
Methods For Microwave Ovens In the U.S. CPI," August 2000;
http://www.bls.gov/cpimwo.htm.
- Specifically, see http://www.applianceadvisor.com/brandadvisor.htm.
- Some information from the ApplianceAdvisor.com website supported the placement of
individual brands into either the "medium" quality or "low" quality
groups.
- Source: "Making Decisions: Buying a Clothes Dryer," Issued March 1998 by the
University of Nebraska in cooperation with the U.S. Department of Agriculture; page 2;
http://www.ianr.unl.edu/pubs/housing/nf348.htm (accessed December 1999).
- Source: "About Clothes Dryer Efficiency," Department of Energy's Office of
Codes and Standards Energy Efficient Appliances; Updated May 24, 2000;
http://www.eren.doe.gov/buildings/consumer_information/dryers/dryabout.html (accessed June
2000).
- See Walter Lane, "Addressing the New Goods Problem in the Consumer Price
Index," Presented at the Issues in Measuring Price Change and Consumption Conference,
Bureau of Labor Statistics, Washington, D.C., June 5-8, 2000, pages 1-26.
References
Consumer Reports
August 2000 Issue, pages 36-38.
Silver, M.S. (1998), "An evaluation of the use of hedonic regressions for basic
components of consumer price indices," Third Meeting of the International Working
Group on Price Indices, Statistics Netherlands, Voorburg: Netherlands (1998) 1-12.
http://www.statcan.ca/secure/english/ottawagroup/pdf/23SIL3.pdf(accessed
November 1999).
Lane, Walter (2000), "Addressing the New Goods Problem in the Consumer Price
Index," Presented at the Issues in Measuring Price Change and Consumption Conference,
Bureau of Labor Statistics, Washington, D.C., June 5-8, 2000, pages 1-26.
Attachment 1
Index levels for HK01 with and without hedonic QAs for clothes dryers
(ELI HK012-cluster 02A)
![](https://webarchive.library.unt.edu/web/20120927123103im_/http://www.bls.gov/cpi/cpidry01.gif)
Attachment 2
Number and Distribution of CPI Price Changes for Major Appliances (HK01)
and
Clothes Dryers (HK012 - 02A)
Major Appliances (HK01) |
|
|
|
|
Month |
Total Prices Collected
|
Substitutions # / (% of Total Prices)
|
Comparable # / (% of Subs)
|
Non-comparable # / (% of Subs)
|
October 1999 |
233 |
33 / (14) |
30 / (91) |
03 / (09) |
November 1999 |
208 |
16 / (08) |
11 / (69) |
05 / (31) |
December 1999 |
209 |
15 / (11) |
09 / (60) |
06 / (40) |
January 2000 |
213 |
22 / (10) |
17 / (77) |
05 / (33) |
February 2000 |
208 |
08 / (04) |
07 / (88) |
01 / (12) |
March 2000 |
210 |
19 / (09) |
14 / (74) |
05 / (26) |
April 2000 |
209 |
20 / (10) |
19 / (95) |
01 / (05) |
May 2000 |
231 |
18 / (08) |
07 / (39) |
11 / (61) |
June 2000 |
245 |
16 / (07) |
15 / (94) |
01 / (06) |
|
|
|
|
|
Average |
218 |
19 / (09) |
15 / (79) |
04 / (21) |
Clothes Dryers (HK012 - 02A) |
|
|
|
|
Month |
Total Prices Collected
|
Substitutions # / (% of Total Prices)
|
Comparable # / (% of Subs)
|
Non-comparable # / (% of Subs)
|
October 1999 |
20 |
03 / (15) |
03 / (100) |
0 |
November 1999 |
29 |
01 / (03) |
01 / (100) |
0 |
December 1999 |
33 |
03 / (09) |
02 / ( 67) |
01 / (33) |
January 2000 |
32 |
06 / (19) |
06 / (100) |
0 |
February 2000 |
34 |
0 |
0 |
0 |
March 2000 |
30 |
02 / (07) |
02 / (100) |
0 |
April 2000 |
36 |
02 / (06) |
02 / (100) |
0 |
May 2000 |
33 |
01 / (03) |
01 / (100) |
0 |
June 2000 |
43 |
02 / (05) |
02 / (100) |
0 |
|
|
|
|
|
Average |
32 |
2.2 / (07) |
2.1 / (95) |
0.1 (05) |
Attachment 3
![](https://webarchive.library.unt.edu/web/20120927123103im_/http://www.bls.gov/cpi/cpidry02.gif)
U.S. Factory Shipments of Clothes Dryers
Data Source: Association of Home Appliance Manufacturers
Attachment 4
Brief Chronology of Clothes Dryer Improvements
- 1930 Electric clothes dryer developed by J. Ross Moore.
- 1936 J. Ross Moore patented clothes dryer which blew hot air through clothes rotating in
a drum.
- 1937 Moore develops gas and electric clothes dryers.
- 1938 Electric dryer put on the market.
- 1946 Controls moved to the front of the clothes dryer.
- 1946 Cool down period added to drying cycle.
- 1946 Sixty-minute timer introduced.
- 1946 Exhaust system introduced which removed moist air from drying room.
- 1946 Temperature controls introduced.
- 1949 Automatic electric dryer introduced.
- 1951 "Moistureless" dryer introduced that carried off moisture and lint
through absorption and condensation.
- 1958 Clothes dryers (30" width) introduced that use a negative pressure system,
which is still in use today.
- 1959 Automatic dryness controls introduced, terminating power when load was dry.
- 1965 Permanent press cycle is introduced.
- 1972 Electric starter on gas clothes dryer introduced.
- 1973 Air condensation dryer introduced, which eliminated need for outside ventilation.
- 1974 Microelectronic controls to time drying cycle introduced.
- 1983 Dryer timer introduced that allowed delayed start, to take advantage of off-peak
rates.
- 1984 Dehumidification dryer introduced, which worked like a home refrigerator.
- 1985 Magnetron invented for use in electric clothes dryer available for licensing.
- 1985 Clothes dryer introduced with all Spanish graphics for console, labels and customer
instructions.
- 1985 Clothes dryer introduced with oversized controls, large graphics on console and big
type instructions.
Data Source: Association of Home Appliance Manufacturers
Attachment 5
Development of Hedonic Model for Clothes
Dryers (HK012 - 02A)
Iterative regression 1
Variable Category
|
Variable Name |
Parameter Estimate |
T Statistic |
Tolerance |
|
Intercept |
4.373587 |
28.156 |
|
Type of Price: |
Sale Price |
-0.112342 |
-4.794 |
0.9702896 |
Dryer Type: |
Gas |
0.270701 |
10.573 |
0.8492123 |
Electric |
Base |
|
|
Drum Capacity (in cubic feet) |
Capacity (in cubic feet) |
0.260096 |
10.994 |
0.8461047 |
R2=0.5396 |
Adjusted R2=0.5356 |
F value=132.07 |
Number of observations=341 |
Iterative regression 2
Variable Category
|
Variable Name |
Parameter Estimate |
T Statistic |
Tolerance |
|
Intercept |
5.482169 |
34.588 |
|
Type of Price: |
Sale Price |
-0.122285 |
-6.245 |
0.9685990 |
Dryer Type: |
Gas |
0.106603 |
4.218 |
0.6073017 |
Electric |
Base |
|
|
Drum Capacity (in cubic feet) |
Capacity (in cubic feet) |
0.030452 |
1.115 |
0.4421581 |
Number of Dry Cycles |
Number of dry cycles |
0.059726 |
12.164 |
0.3249735 |
R2=0.6801 |
Adjusted R2=0.6763 |
F value=179.11 |
Number of observations=341 |
Iterative regression 3
Variable Category
|
Variable Name |
Parameter Estimate |
T Statistic |
Tolerance |
|
Intercept |
5.001985 |
36.693 |
|
Type of Price: |
Sale Price |
-0.115304 |
-6.003 |
0.9701509 |
Dryer Type: |
Gas |
0.187317 |
8.531 |
0.7755253 |
Electric |
Base |
|
|
Drum Capacity (in cubic feet) |
Capacity (in cubic feet) |
0.148203 |
6.976 |
0.7048710 |
Drying Mechanism |
Moisture sensor |
0.288525 |
12.891 |
0.6808047 |
Temp sensor / Timed dry |
Base |
|
|
R2=0.6917 |
Adjusted R2=0.6880 |
F value=189.01 |
Number of observations=341 |
Iterative regression 4
Variable Category
|
Variable Name |
Parameter Estimate |
T Statistic |
Tolerance |
|
Intercept |
4.600270 |
33.178 |
|
Type of Price: |
Sale Price |
-0.090030 |
-4.937 |
0.9344482 |
Dryer Type: |
Gas |
0.203266 |
9.878 |
0.7665995 |
Electric |
Base |
|
|
Drum Capacity (in cubic feet) |
Capacity (in cubic feet) |
0.188446 |
9.164 |
0.6531452 |
Drying Mechanism |
Moisture sensor |
0.170382 |
6.429 |
0.4215581 |
Temp sensor / Timed dry |
Base |
|
|
Number of Temperature Settings |
Number of temperatures (per temperature) |
0.038153 |
7.224 |
0.5952816 |
R2=0.7331 |
Adjusted R2=0.7292 |
F value=184.61 |
Number of observations=341 |
Iterative regression 5
Variable Category
|
Variable Name |
Parameter Estimate |
T Statistic |
Tolerance |
|
Intercept |
4.660102 |
34.195 |
|
Type of Price: |
Sale Price |
-0.084458 |
-4.725 |
0.9290540 |
Dryer Type: |
Gas |
0.197730 |
9.811 |
0.7631638 |
Electric |
Base |
|
|
Drum Capacity (in cubic feet) |
Capacity (in cubic feet) |
0.178732 |
8.833 |
0.6442004 |
Drying Mechanism |
Moisture sensor |
0.154114 |
5.882 |
0.4119027 |
Temp sensor / Timed dry |
Base |
|
|
Number of Temperature Settings |
Number of temperatures (per temperature) |
0.038580 |
7.474 |
0.5950391 |
Dryer Control Types |
Electronic controls |
0.124248 |
4.103 |
0.8796670 |
Rotary / Push button controls |
Base |
|
|
R2=0.7459 |
Adjusted R2=0.7413 |
F value=163.89 |
Number of observations=341 |
Iterative regression 6
Variable Category
|
Variable Name |
Parameter Estimate |
T Statistic |
Tolerance |
|
Intercept |
5.253683 |
35.531 |
|
Type of Price: |
Sale Price |
-0.096046 |
-6.414 |
0.8555140 |
Dryer Type: |
Gas |
0.175961 |
10.058 |
0.6546856 |
Electric |
Base |
|
|
Drum Capacity (in cubic feet) |
Capacity (in cubic feet) |
0.118828 |
5.528 |
0.3690016 |
Drying Mechanism |
Moisture sensor |
0.080280 |
3.325 |
0.3134690 |
Temp sensor / Timed dry |
Base |
|
|
Number of Temperature Settings |
Number of temperatures (per temperature) |
0.026065 |
5.660 |
0.4833153 |
Dryer Control Types |
Electronic controls |
0.182129 |
7.259 |
0.8284481 |
Rotary / Push button controls |
Base |
|
|
Brand: |
Spec B1 |
-0.294900 |
-6.698 |
0.8262573 |
Spec B2 |
-0.194441 |
-4.202 |
0.8286723 |
Spec B3 |
-0.080824 |
-1.070 |
0.9169240 |
Spec B4 |
-0.303760 |
-8.041 |
0.5786202 |
Spec B5 |
-0.187714 |
-6.068 |
0.6321185 |
Spec B7 |
-0.342112 |
-6.161 |
0.8554288 |
Spec B11 |
0.007234 |
0.155 |
0.8117872 |
Spec B13 |
0.007500 |
0.282 |
0.4846242 |
Spec B15 |
-0.348871 |
-4.631 |
0.9213799 |
Spec B16 |
-0.218829 |
-8.971 |
0.6333509 |
Other brands not listed |
Base |
|
|
R2=0.8407 |
Adjusted R2=0.8328 |
F value=107.16 |
Number of observations=341 |
Iterative regression 7
Variable Category
|
Variable Name |
Parameter Estimate |
T Statistic |
Tolerance |
|
Intercept |
5.438312 |
38.914 |
|
Type of Price: |
Sale Price |
-0.101541 |
-7.007 |
0.8811153 |
Dryer Type: |
Gas |
0.164389 |
9.825 |
0.6904156 |
Electric |
Base |
|
|
Drum Capacity (in cubic feet) |
Capacity (in cubic feet) |
0.083562 |
3.956 |
0.3684787 |
Drying Mechanism |
Moisture sensor |
0.078286 |
3.564 |
0.3652782 |
Temp sensor / Timed dry |
Base |
|
|
Number of Temperature Settings |
Number of temperatures (per temperature) |
0.029538 |
6.879 |
0.5362365 |
Dryer Control Types |
Electronic controls |
0.182907 |
7.432 |
0.8303869 |
Rotary / Push button controls |
Base |
|
|
Brand: |
Spec B1 |
-0.237387 |
-5.240 |
0.7528344 |
Spec B2 |
-0.133392 |
-2.806 |
0.7572554 |
Spec B4 |
-0.290021 |
-8.431 |
0.6730128 |
Spec B5 |
-0.158421 |
-5.438 |
0.6874771 |
Spec B7 |
-0.338281 |
-6.347 |
0.8956869 |
Spec B15 |
-0.347624 |
-4.764 |
0.9471823 |
Spec B16 |
-0.230276 |
-10.521 |
0.7586572 |
Other brands not listed |
Base |
|
|
Door Style: |
Front-pull |
0.070078 |
3.369 |
0.4110697 |
Side-pull |
Base |
|
|
R2=0.8454 |
Adjusted R2=0.8387 |
F value=127.68 |
Number of observations=341 |
Iterative regression 8
Variable Category
|
Variable Name |
Parameter Estimate |
T Statistic |
Tolerance |
|
Intercept |
5.665578 |
37.395 |
|
Type of Price: |
Sale Price |
-0.089546 |
-6.455 |
0.8596389 |
Dryer Type: |
Gas |
0.146698 |
8.000 |
0.5137997 |
Electric |
Base |
|
|
Drum Capacity (in cubic feet) |
Capacity (in cubic feet) |
0.038984 |
1.689 |
0.2759290 |
Drying Mechanism |
Moisture sensor |
0.040024 |
1.627 |
0.2603760 |
Temp sensor / Timed dry |
Base |
|
|
Number of Temperature Settings |
Number of temperatures (per temperature) |
0.020085 |
4.529 |
0.4493292 |
Dryer Control Types |
Electronic controls |
0.171299 |
4.673 |
0.3345893 |
Rotary / Push button controls |
Base |
|
|
Brand: |
Spec B1 |
-0.329589 |
-7.643 |
0.7426032 |
Spec B2 |
-0.115597 |
-2.579 |
0.7618037 |
Spec B4 |
-0.293926 |
-8.433 |
0.5860779 |
Spec B5 |
-0.170593 |
-6.179 |
0.6842013 |
Spec B7 |
-0.250686 |
-4.744 |
0.8144487 |
Spec B15 |
-0.224560 |
-3.118 |
0.8692683 |
Spec B16 |
-0.155692 |
-5.988 |
0.4806630 |
Other brands not listed |
Base |
|
|
Features: |
Drying rack |
0.070079 |
3.019 |
0.3220153 |
Drum light |
0.098307 |
4.115 |
0.4132451 |
Easy access lint filter |
-0.070600 |
-3.930 |
0.5872031 |
Lint filter indicator |
0.011447 |
0.309 |
0.3277456 |
Leveling legs |
0.048375 |
2.183 |
0.8688682 |
Sound insulation |
0.004626 |
0.201 |
0.3111389 |
End of cycle signal |
0.050141 |
1.973 |
0.40911523 |
R2=0.8643 |
Adjusted R2=0.8558 |
F value=102.22 |
Number of observations=341 |
Iterative regression 9
Variable Category
|
Variable Name |
Parameter Estimate |
T Statistic |
Tolerance |
|
Intercept |
5.667737 |
37.595 |
|
Type of Price: |
Sale Price |
-0.090446 |
-6.650 |
0.8890915 |
Dryer Type: |
Gas |
0.146596 |
8.042 |
0.5169445 |
Electric |
Base |
|
|
Drum Capacity (in cubic feet) |
Capacity (in cubic feet) |
0.038942 |
1.692 |
0.2759917 |
Drying Mechanism |
Moisture sensor |
0.040517 |
1.833 |
0.3207535 |
Temp sensor / Timed dry |
Base |
|
|
Number of Temperature Settings |
Number of temperatures (per temperature) |
0.020218 |
4.596 |
0.4541020 |
Dryer Control Types |
Electronic controls |
0.180106 |
7.671 |
0.8109550 |
Rotary / Push button controls |
Base |
|
|
Brand: |
Spec B1 |
-0.332106 |
-8.065 |
0.8096925 |
Spec B2 |
-0.117618 |
-2.731 |
0.8198770 |
Spec B4 |
-0.296335 |
-8.974 |
0.6491349 |
Spec B5 |
-0.172063 |
-6.332 |
0.7022082 |
Spec B7 |
-0.253148 |
-4.858 |
0.8326990 |
Spec B15 |
-0.228218 |
-3.220 |
0.8921679 |
Spec B16 |
-0.159437 |
-6.763 |
0.5811767 |
Other brands not listed |
Base |
|
|
Features: |
Drying rack |
0.071850 |
3.219 |
0.3462451 |
Drum light |
0.097172 |
4.157 |
0.4292123 |
Easy access lint filter |
-0.069691 |
-3.963 |
0.6092316 |
Leveling legs |
0.048523 |
2.199 |
0.8714779 |
End of cycle signal |
0.052042 |
2.230 |
0.4821264 |
R2=0.8642 |
Adjusted R2=0.8567 |
F value=114.22 |
Number of observations=341 |
Iterative regression 10
Variable Category
|
Variable Name |
Parameter Estimate |
T Statistic |
Tolerance |
|
Intercept |
5.259427 |
37.968 |
|
Type of Price: |
Sale Price |
-0.093481 |
-6.533 |
0.8844529 |
Dryer Type: |
Gas |
0.165436 |
9.844 |
0.6696858 |
Electric |
Base |
|
|
Drum Capacity (in cubic feet) |
Capacity (in cubic feet) |
0.111995 |
5.699 |
0.4167259 |
Drying Mechanism |
Moisture sensor |
0.082280 |
3.725 |
0.3536422 |
Temp sensor / Timed dry |
Base |
|
|
Number of Temperature Settings |
Number of temperatures (per temperature) |
0.023868 |
5.599 |
0.5323425 |
Dryer Control Types |
Electronic controls |
0.180812 |
7.402 |
0.8249734 |
Rotary / Push button controls |
Base |
|
|
Brand: |
Spec B1 |
-0.317507 |
-7.494 |
0.8422399 |
Spec B2 |
-0.181093 |
-3.677 |
0.6903681 |
Spec B4 |
-0.297064 |
-8.575 |
0.6495386 |
Spec B5 |
-0.175545 |
-6.294 |
0.7339691 |
Spec B7 |
-0.310267 |
-5.816 |
0.8750382 |
Spec B15 |
-0.283702 |
-3.850 |
0.9089035 |
Spec B16 |
-0.178679 |
-7.492 |
0.6253064 |
Other brands not listed |
Base |
|
|
Features: |
Easy access lint filter |
-0.065990 |
-3.605 |
0.6190762 |
Leveling legs |
0.070103 |
3.088 |
0.9062395 |
Warranty: |
2 years parts and labor |
0.030173 |
0.739 |
0.6558679 |
2 years parts / 1 year labor |
0.001180 |
0.042 |
0.7774035 |
1 year parts / 2 years labor |
0.039395 |
0.322 |
0.9810179 |
1 year parts and labor |
Base |
|
|
R2=0.8505 |
Adjusted R2=0.8422 |
F value=102.09 |
Number of observations=341 |
Iterative regression 11
Variable Category
|
Variable Name |
Parameter Estimate |
T Statistic |
Tolerance |
|
Intercept |
5.319622 |
37.592 |
|
Type of Price: |
Sale Price |
-0.106089 |
-7.105 |
0.7933756 |
Dryer Type: |
Gas |
0.155015 |
9.155 |
0.6445137 |
Electric |
Base |
|
|
Drum Capacity (in cubic feet) |
Capacity (in cubic feet) |
0.107894 |
5.450 |
0.4011680 |
Drying Mechanism |
Moisture sensor |
0.082778 |
3.738 |
0.3436128 |
Temp sensor / Timed dry |
Base |
|
|
Number of Temperature Settings |
Number of temperatures (per temperature) |
0.023831 |
5.613 |
0.5243144 |
Dryer Control Types |
Electronic controls |
0.149904 |
3.980 |
0.3389662 |
Rotary / Push button controls |
Base |
|
|
Brand: |
Spec B1 |
-0.319781 |
-7.560 |
0.8254853 |
Spec B2 |
-0.149847 |
-3.170 |
0.7323834 |
Spec B4 |
-0.279947 |
-8.070 |
0.6328471 |
Spec B5 |
-0.173011 |
-5.776 |
0.6216921 |
Spec B7 |
-0.297178 |
-5.565 |
0.8529136 |
Spec B15 |
-0.275868 |
-3.724 |
0.8786021 |
Spec B16 |
-0.155726 |
-6.155 |
0.5428946 |
Other brands not listed |
Base |
|
|
Features: |
Easy access lint filter |
-0.056874 |
-3.172 |
0.6302839 |
Leveling legs |
0.056268 |
2.361 |
0.8036933 |
Country of Origin: |
Canada |
0.012862 |
0.225 |
0.8931072 |
USA |
Base |
|
|
Color: |
Almond |
0.023602 |
0.608 |
0.8941839 |
Graphite |
0.041724 |
0.961 |
0.3543941 |
White |
Base |
|
|
Type of Outlet: |
Furniture |
0.004664 |
0.125 |
0.7729034 |
Appliance |
-0.025863 |
-1.272 |
0.4844347 |
Electronics |
-0.052913 |
-1.627 |
0.7558829 |
Hardware |
-0.116291 |
-2.872 |
0.7551572 |
Warehouse |
-0.258230 |
-2.819 |
0.8592501 |
Region / City Size: |
Northeast |
-0.011941 |
-0.551 |
0.9810179 |
Midwest |
-0.005378 |
-0.294 |
0.7043850 |
West |
-0.005205 |
-0.295 |
0.6912077 |
A-Size |
0.004951 |
0.348 |
0.8390525 |
C-Size |
-0.018608 |
-0.322 |
0.8743021 |
R2=0.8585 |
Adjusted R2=0.8458 |
F value=67.81 |
Number of observations=341 |
Iterative regression 12
Variable Category
|
Variable Name |
Parameter Estimate |
T Statistic |
Tolerance |
|
Intercept |
5.304425 |
40.649 |
|
Type of Price: |
Sale Price |
-0.104647 |
-7.248 |
0.8714113 |
Dryer Type: |
Gas |
0.171883 |
10.580 |
0.7188645 |
Electric |
Base |
|
|
Drum Capacity (in cubic feet) |
Capacity (in cubic feet) |
0.112726 |
5.978 |
0.4539854 |
Drying Mechanism |
Moisture sensor |
0.084390 |
3.868 |
0.3634881 |
Temp sensor / Timed dry |
Base |
|
|
Number of Temperature Settings |
Number of temperatures (per temperature) |
0.025470 |
6.108 |
0.5579774 |
Dryer Control Types |
Electronic controls |
0.181103 |
7.419 |
0.8285117 |
Rotary / Push button controls |
Base |
|
|
Brand: |
Spec B5 |
-0.195326 |
-7.087 |
0.7537983 |
Spec B2 |
-0.196404 |
-4.470 |
0.8702357 |
Spec B16 |
-0.204300 |
-9.380 |
0.7521143 |
Spec B1 |
-0.294836 |
-7.042 |
0.8650109 |
Spec B4 |
-0.295535 |
-8.707 |
0.6786309 |
Spec B7 |
-0.332486 |
-6.283 |
0.8916961 |
Spec B15 |
-0.356184 |
-4.925 |
0.9465843 |
Other brands not listed |
Base |
|
|
Type of Outlet: |
Hardware |
-0.110182 |
-2.955 |
0.9144920 |
Warehouse |
-0.289297 |
-3.281 |
0.9524403 |
R2=0.8487 |
Adjusted R2=0.8417 |
F value=121.88 |
Number of observations=341 |
F Test for Brand Equivalence
Test 1: H0 : Spec B5 = Spec B2 = Spec B16
Numerator: 0.5775 |
DF: 3 |
F value: 39.2283 |
Denominator: 0.014721 |
DF: 326 |
Prob>F: 0.0001 |
F Test 1 fails to reject H0
Test 2: H1 : Spec B1 = Spec B4 = Spec B7 = Spec B15
Numerator: 0.5360 |
DF: 4 |
F value: 36.4118 |
Denominator: 0.014721 |
DF: 326 |
Prob>F: 0.0001 |
F Test 2 fails to reject H1
Final regression
Variable Category
|
Variable Name |
Parameter Estimate |
T Statistic |
Tolerance |
|
Intercept |
5.295815 |
46.235 |
|
Type of Price: |
Sale Price |
-0.103660 |
-7.257 |
0.8797381 |
Dryer Type: |
Gas |
0.170917 |
10.641 |
0.7268179 |
Electric |
Base |
|
|
Drum Capacity (in cubic feet) |
Capacity (in cubic feet) |
0.113937 |
6.943 |
0.5924369 |
Drying Mechanism |
Moisture sensor |
0.082960 |
3.920 |
0.3817178 |
Temp sensor / Timed dry |
Base |
|
|
Number of Temperature Settings |
Number of temperatures (per temperature) |
0.025726 |
6.255 |
0.5668198 |
Dryer Control Types |
Electronic controls |
0.181387 |
7.544 |
0.8441393 |
Rotary / Push button controls |
Base |
|
|
Brand: |
Low quality |
-0.305682 |
-12.153 |
0.6655662 |
Medium quality |
-0.200536 |
-10.915 |
0.6747474 |
High quality |
Base |
|
|
Type of Outlet: |
Hardware |
-0.110854 |
-3.038 |
0.9439328 |
Warehouse |
-0.292446 |
-3.366 |
0.9692976 |
R2=0.8481 |
Adjusted R2=0.8436 |
F value=184.87 |
Number of observations=341 |
Attachment 6
1999-2000 U.S. Level Price Relatives, Indexes and
Index Percent Changes for CPI Item SEHK01, Major Appliances
US level Item SEHK01, Major Appliances
Clothes Dryers (HK012 - 02A) quality adjustments
|
|
Without
Quality Adjustments |
|
With
Quality Adjustments |
Month |
|
Price Relatives |
Indexes |
1 Month
% Change |
|
Price Relatives |
Indexes |
1 Month
% Change |
|
|
|
|
|
|
September 1999 |
|
98.119 |
|
|
98.119 |
|
October 1999 |
0.99630 |
97.756 |
-0.37 |
0.99601 |
97.728 |
-0.40 |
November 1999 |
1.00807 |
98.545 |
0.81 |
1.00802 |
98.511 |
0.80 |
December 1999 |
0.99816 |
98.364 |
-0.18 |
0.99847 |
98.361 |
-0.15 |
January 2000 |
0.99239 |
97.615 |
-0.76 |
0.99288 |
97.660 |
-0.71 |
February 2000 |
0.99923 |
97.540 |
-0.08 |
0.99923 |
97.585 |
-0.08 |
March 2000 |
1.00324 |
97.856 |
0.32 |
1.00325 |
97.902 |
0.32 |
April 2000 |
1.00319 |
98.168 |
0.32 |
1.00320 |
98.215 |
0.32 |
May 2000 |
0.99569 |
97.745 |
-0.43 |
0.99569 |
97.792 |
-0.43 |
June 2000 |
0.99949 |
97.695 |
-0.05 |
0.99948 |
97.741 |
-0.05 |
Sep to Jun Oct to Jun |
-0.43
-0.06 |
-0.39
0.01 |
Attachment 7
![](https://webarchive.library.unt.edu/web/20120927123103im_/http://www.bls.gov/cpi/cpidry03.gif)
Attachment 8
ELI Checklist HK012—Cluster 02A, Dryers
BUREAU OF LABOR STATISTICS U.S. DEPARTMENT OF LABOR
CONSUMER PRICE INDEX - ELI CHECKLIST
collection outlet quote arranging
period: __ __ __ __ number: __ __ __ __ __ __ __ code: __ __ __ code: __ __ __ __
_________________________________________________________________________________________
ELI No./ cluster
title HK012 WASHERS AND DRYERS code 02A
item availability: 1-AVAILABLE 2-ELI NOT SOLD 3-INIT INCOMPLETE
purpose of checklist: 1-INIT 2-INIT COMPL 3-SPEC CORR 4-SUB 5-REINIT 6-CHECK REV
_________________________________________________________________________________________
CURRENT PERIOD | SALES TAX
|
price _ _ _ _ _ _ . _ _ _ | included: YES NO
|
type of price: REG SALE |
|
|
|
|
|
|
|
YEAR-ROUND | in-season: JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC
____________|____________________________________________________________________________
respondent: location:
_________________________________________________________________________________________
field message:
_________________________________________________________________________________________
CLUSTER 02A - DRYERS
TYPE CAPACITY CUBIC FEET
A1 Electric (Super-capacity and models)
A2 Gas D1 7.4 cubic feet or greater
D2 7.0 to 7.3 cubic feet
BRAND (Extra large and large-capacity models)
B1 Admiral D3 6.6 to 6.9 cubic feet
B2 Amana D4 6.2 to 6.5 cubic feet
B3 Crosley D5 5.8 to 6.1 cubic feet
B4 Fridgidaire D6 5.4 to 5.7 cubic feet
B5 General Electric (GE) D7 5.0 to 5.3 cubic feet
B6 Gibson (Compact-capacity models)
B7 Hot Point D8 4.6 to 4.9 cubic feet
B8 Jenn-Air D9 4.2 to 4.5 cubic feet
B9 Kelvinator D10 3.8 to 4.1 cubic feet
B10 Kenmore D11 3.7 cubic feet or less
B11 Kitchenaid
B12 Magic Chef ** SPECIFIC CAPACITY IN CUBIC FEET
B13 Maytag
B14 RCA E99 ______________________________
B15 Roper
B16 Whirlpool
B17 White-Westinghouse
B99 Other brand,
______________________________
** MANUFACTURER'S MODEL NUMBER
C99 ______________________________
ZZ99
_________________________________________________________________________________________
HK012-02A - DRYERS - CONTINUED
CYCLES(Must agree with next
DOOR STYLE
specification category) AE1 Front-pull (opens down)
F1 Regular AE2 Side-pull (opens left or right)
G1 Permanent press
H1 Gentle/knits/delicates
FEATURES
I1 Air fluff/no heat AF1 Drying rack
J1 Damp dry AG1 Drum/interior light
K1 Easy/extra care AH1 Easy access lint filter on dryer top
L1 Wrinkle free/wrinkle release AI1 Lint filter indicator
M1 Press care/wrinkle prevent AJ1 Leveling legs
N99 Other dry cycle, AK1 Sound dampening package
AL1 End of cycle signal
______________________________ AM1 Stackable unit
P99 Other dry cycle, AN99 Other features,
______________________________ ______________________________
Q99 Other dry cycle, AP1 None
______________________________
MANUFACTURER WARRANTY
AQ1 One year parts and labor
NUMBER OF DRY CYCLES(Must agree with AQ99 Other warranty,
previous specification category)
R99 Specific number of dry cycles, ______________________________
______________________________
COUNTRY OF ORIGIN
AR1 United States
DRYING MECHANISMS AR99 Other country of origin,
S1 Automatic moisture sensor
(electronic sensor) ______________________________
T1 Automatic temperature sensor
U1 Timed drying
DELIVERY
V99 Other, AS1 Delivery available, no extra charge
AS2 No delivery available
______________________________ AS79 Delivery available for extra charge,
NUMBER OF TEMPERATURE SETTINGS __________________// $________
W1 Two settings
W2 Three settings
COLOR
W3 Four settings AT1 White
W4 Five settings AT2 Almond
W5 Six settings AT99 Other color,
W6 Seven settings
W7 Variable or infinite settings ______________________________
W99 Other number of temperature AU1 No extra charge for color
settings, collected in AT spec
AU79 Extra charge for color,
______________________________
__________________// $________
CONTROL TYPES
X1 Touch pad or electronic controls
Y1 Rotary controls
AA1 Push button controls
AB99 Other control types,
______________________________
NUMBER OF MOTOR SPEEDS
AC1 One speed
AC2 Two speeds
AC3 Three speeds
AC99 Other number of motor speeds
______________________________
DRYING DRUM MATERIAL
AD1 Porcelain enamel drum
AD2 Stainless steel drum
AD99 Other drum material,
______________________________
HK012-02A - DRYERS - CONTINUED
OTHER PRICE FACTORS
AV99 ______________________________
AW99 ______________________________
AX99 ______________________________
SERIES OR LINE NAME (IF APPLICABLE)
AY99 ______________________________
**OTHER CLARIFYING INFORMATION
BA99 ______________________________
BB99 ______________________________
BC99 ______________________________
PRICE CALCULATION BOX
BD79 Dryer base price, ______________________________// $__________
BE89 TOTAL PRICE (AS79 + AU79 + BD79), ____________________________// $__________
Last Modified Date: April 22, 2003