This paper estimates the hedonic price function from January 1996 to October 2019 using POS data of TVs, air conditioners, refrigerators, rice cookers, vacuum cleaners, and washing machines. While controlling for the standard specifications of each product, POS data was used to observe whether the degree of energy-saving functions and the presence or absence of additional functions contributed to higher value added. The largest difficulty is determining what specifications are representative of each product over an approximate 20-year span. We have constructed a new dataset that dataset for which this 20-year analysis is possible. In the estimation results, we were able to observe the effects of past and current consumption tax increases on each product market, as well as the effects of extreme weather, such as high temperatures, on each product market. We also observed technological changes in the home appliance market in Japan over the last 20 years, by counting the number of individual functions that home appliances featured and comparing them over time. Our estimation results showed that product size / volume and energy-saving functions are effective in explaining the price model of home appliances.