The residential sector uses a large fraction of energy for heating and cooling to provide comfort conditions for households. Technical features of appliances are determining factors of the demand for energy. The present paper thus aims to study the determinants of household adoption decision by applying the discrete choice model. Households face with unordered choice, and multinomial and mixed logit models are employed to study the heterogeneity of choices of households related to heating and cooling technologies. While most of the research works have considered preferences based on cross-sectional data collected through discrete choice experiments, the present study employs micro-level household survey data including observed preferences in order to contribute to better understanding of the actual choice behavior of households. In addition, the dynamic effect of variation of variables on the household preferences have been investigated over time. The results show that heterogenous preferences of families in choosing appliances are affected by socio-demographic characteristics. Analysis of marginal effects indicate that urbanization, education, building type and households expenditure are the most effective parameters. The level of education has the most positive impact on the adoption of wall-mounted package and split air conditioner where the probability of using them has increased by 1.65% and 4.87%, respectively. On the other hand, high education level reduces the probability of gas heater and fan adoption by 3.49% and 3.43%. The estimated results show that urbanization increases the probability of choosing central heating by 3.73% and split air conditioner by 6.1% while it decreases the adoption probability of oil heater by 5.02% and fan as by 8.9%.
Read full abstract