Department of Labor Logo United States Department of Labor
Dot gov

The .gov means it's official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you're on a federal government site.


The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Consumer Price Index
Microwaves image

Item definition

The Entry Level Item (ELI) category of Microwave Ovens is part of the index for Major Appliances, and includes all microwave ovens that are not part of a range. In this factsheet, we examine the regression model used in the quality adjustment of price data for microwave ovens.

CPI sample

The sample used for the new hedonic model included approximately 480 observations. About 80 percent of observations were sourced from data provided by an appliance retailer, while the remaining 20 percent were quotes from traditional collection for the CPI. All prices were from August 2022, and there were 45 brands in the sample and more than 420 distinct microwave models. The top 3 brands accounted for 28 percent of the observations, and there were 22 distinct outlets in the sample.  

Current Model

Two separate models – one for countertop microwaves and one for over-the-range microwaves accounted for nearly 80 percent of sample observations (two other types of microwaves are priced in the CPI: drawer and built-in styles and were not modelled). The choice to model counter and over-the-range styles separately was made for two primary reasons.

First, the relationship between capacity and price for countertop microwaves is quite linear: the larger the microwave, the higher the price tends to be. However, this relationship for over-the-range microwaves was found to be “V” shaped since some models with the least capacity had the highest prices and were low profile design. Low-profile, over-the-range microwaves are about 10” in height, compared to 16” for a standard size, and are designed to fit under kitchen cabinets or above ranges where space is limited. And, secondly, over-the-range microwaves are capable of exhausting smoke and odors from the cooking space, a feature that is not present in the countertop style.

Table 1 presents the details of the current model used to quality adjust countertop microwaves.

Table 1. Parameter estimates for the hedonic regression for countertop microwaves
Variable DF Parameter Estimate Standard Error t Value Pr > |t| Tolerance


1 4.4372 0.07895 56.21 <.0001

Heating method: Convection

1 0.69347 0.06205 11.18 <.0001 0.81301

Size: Capacity

1 0.4071 0.4071 6.57 <.0001 0.433

Brand: #1

1 -0.39841 0.09992 -3.99 0.0001 0.8797

Brand: #2

1 0.20404 0.07456 2.74 0.007 0.83153

Brand: #3

1 0.23575 0.07048 3.34 0.0011 0.87511

Brand: #4

1 0.79609 0.11093 7.89 <.0001 0.86223

Brand: Luxury

1 1.08201 0.07887 13.72 <.0001 0.79427

Manufacturer’s Series: In a line or series

1 0.51035 0.09599 5.32 <.0001 0.75207

Features: Sensor cooking

1 0.1483 0.06026 2.46 0.0151 0.44219

Features: 7+ programs

1 0.21781 0.05211 4.18 <.0001 0.70885

Control: Discount Outlet #1

1 -0.58844 0.0846 -6.96 <.0001 0.80392

Table 2 presents the details of the current model used to quality adjust over-the-range microwaves.

Table 2. Parameter estimates for the hedonic regression for over-the-range microwaves
Variable DF Parameter Estimate Standard Error t Value Pr > |t| Tolerance


1 3.13778 0.48788 6.43 <.0001

Heating method: Convection

1 0.32221 0.03038 10.6 <.0001 0.63754

Color: Black or white

1 -0.15017 -0.15017 -6.06 <.0001 0.82325

Color: Black stainless steel

1 0.08704 0.0295 2.95 0.0035 0.8598

Design: Low profile

1 0.16558 0.04487 3.69 0.0003 0.68097

Brand: #1

1 -0.51328 0.15264 -3.36 0.0009 0.98172

Brand: #2

1 -0.17235 0.03077 -5.6 <.0001 0.84906

Brand: #3

1 0.06563 0.03149 2.08 0.0383 0.75459

Brand: #4

1 0.43584 0.04948 8.81 <.0001 0.70689


1 0.63585 0.04343 14.64 <.0001 0.76606

Manufacturer’s Series: In a line or series

1 0.20742 0.03167 6.55 <.0001 0.74596

Features: “Smart” capability

1 0.13244 0.03953 3.35 <.0001 0.73138

Features: Sensor cooking

1 0.17972 0.02607 6.89 <.0001 0.65863

Ventilation: Change in fan rating (CFM)

1 0.47733 0.08532 5.59 <.0001 0.52323
  • Quality adjustments derived from hedonic regressions for microwaves were first used in the CPI in 2000. Since then, the model for microwaves has been updated 2 times, including the August 2022 update described in this report.

  • Parameters that were statistically significant in each model above include convection heating method, sensor cooking feature, and models in a manufacturer’s series. A convection microwave uses a heating element and a fan to distribute heat throughout the appliance. This feature enables the appliance to bake and roast food like a traditional oven. Microwaves with sensor cooking detects the amount of steam emitted during the heating process for more precise cooking. Appliances in a manufacturer’s line or series include, for example, microwaves in General Electric’s Profile or Frigidaire’s Gallery series.

  • Variables that were statistically significant in only the countertops model include capacity and a high (more than 6) number or pre-set programs. A microwave’s number of pre-set cooking programs is, at times, hard to determine. Most models have buttons for cooking a beverage, potato, popcorn, and vegetable, for example. Others have additional options that are not so apparent. A LOESS (locally estimated scatterplot smoothing) curve for countertop models revealed that prices increase for units with more than 6 programs.

  • Variables that were significant in only the over-the-range model include color, low profile design, “smart” features, and exhaust fan rating. The a priori for why color was significant for only the over-the-range microwaves is the desire to color coordinate appliances with other kitchen finishes. Smart features include Wi-Fi connectivity or the ability to pair with a manufacturer’s phone application for troubleshooting or downloading recipes. The exhaust fan rating parameter is applied as the log in the change in one microwave’s rating, measured in cubic feet per minute, or CFM, to another model’s rating.

  • Characteristics that were found to not have statistically significant effects on price in either model include maximum cooking wattage, fingerprint-resistant finish, child lock feature, and the presence of a rotating turntable.

  • The final hedonic models in Tables 1 and 2 above were selected using a machine learning technique called a k-folds cross validation. Each sample was divided into 10 parts. Interim models were created in an iterative process using 90 percent of observations as the training dataset and 10 percent as the test dataset. The model selected was the one with the lowest prediction error.

  • The R2 statistics were .906 for the countertops model and 0.872 for the over-the-range model. These values indicate that approximately 90 percent of the variation in price are explained by each hedonic regression model.


Access data for Major appliances in our online database.

Additional information

Additional information may be obtained from the Consumer Price Index Information Office by email or calling 202-691-7000. Information on the CPI's overall methodology can be found in the BLS Handbook of Methods.

Last Modified Date: April 5, 2024