Modeling Ontario regional electricity system demand using a mixed fixed and random coefficients approach

Panel data analysis—advantages and challenges

“…If individual behaviors are similar, conditional on certain variables, panel data provide the possibility of learning an individual's behavior by observing the behavior of others. Thus, it is possible to obtain a more accurate description of an individual's behavior by supplementing observations of the individual in question with data on other individuals (e.g., Hsiao et al 1993Hsiao et al , 1989 Granger 1990;Lewbel 1994;Pesaran 2003), but policy evaluation based on aggregate data may be grossly misleading. Furthermore, the prediction of aggregate outcomes using aggregate data can be less accurate than the prediction based on micro-equations (e.g., .…”

Electricity demand in Kazakhstan

“…Many of the reduced-form electricity demand models that analyse industrial, residential, and service sectors in both developed and developing countries rely upon the dynamic partial adjustment model of Balestra and Nerlove (1966), Houthakker and Taylor (1970), andHouthakker et al (1974) (e.g., Berndt andSamaniego, 1984;Hsaio et al, 1989;Diabi, 1998;Liu, 2004;Bigano et al, 2006). The partial adjustment model relies on assumptions of capital stock as a determinant of the level and the growth rate of electricity demand and the degree of substitution flexibility.…”

Modeling demand structure using scanner data: Implications for salmon enhancement policies

“…However, the SUR approach treats the parameters as fixed, while the random coefficients model treats regional differences as random outcomes from a common population. This approach has been used for other panel studies, such as Hsiao, Mountin, Chan, and Tsui (1989). Since we can test to determine whether or not the coefficients are fixed or random with the random coefficient specification, we choose to estimate the random coefficient model to test parameter stability in the demand for canned salmon.…”