Academic journal article International Advances in Economic Research

Predicting Electricity Consumption in Corporate Sector-Case of the Czech Republic

Academic journal article International Advances in Economic Research

Predicting Electricity Consumption in Corporate Sector-Case of the Czech Republic

Article excerpt

JEL C10 Econometrics * Statistics * E20 Consumption/saving/investment * L50 Regulation and industrial policy * L60 Industry studies * 050 Country studies

This paper is concerned with corporate sector forecasts and it is based on econometric modeling with a use of macro-data. The prediction of electricity demand needs is becoming more difficult throughout Europe, because of the EU incentives to support renewable weather-affected energy sources. Therefore, it is interesting to predict the demand for electricity, which allows us to plan for establishing new energy generators and optimizing energy mix. The electricity market regulator's point of view and knowledge of future electricity consumption trends improves network loss forecasts in electricity transmission and distribution. With high quality loss prediction, the regulator is able to properly estimate, losses and incorporate them into the national electricity tariffs.

The econometric model of electricity consumption in firms describes which exogenous variables influence endogenous electricity consumption using the OLS method. Possible exogenous variables might be the same as in households--income, weather or electricity prices according to Aroonruengsawat et al. (2009). Similarly, Ozttwk and Acaravci (2011) show causalities between economic growth and electricity consumption. All similar papers demonstrate various connections between macroeconomic performance indicators in various data sets and electricity consumption.

Extrapolations of the exogenous variable time series are made using quarterly data and exponential smoothing. Endogenous corporate electricity consumption is explained using a set of 13 exogenous variables from the fields of economics, demography, and meteorology. The best quality results were attained using first differences of the original data (the logarithmic approach was abandoned due to the presence of negative values in the time series). Similarly, some vectors were omitted from the model because of poor quality or methodological disturbances, while some were added with the goal of improving the forecast--the weather vector was multiplied by a unitary matrix creating 4 new vectors representing different consumption behavior in different parts of the year:

[DELTA][y.sub.2] = [[beta].sub.0] + [DELTA][[beta].sub.1]pocx1 + [DELTA][[beta].sub.2]pocr2 + [DELTA][[beta].sub.3]pocx3 + [DELTA][[beta].sub.4]pocv4 + [[beta].sub.4][x.sub.4] + [[beta].sub.5][x.sub.5] + [[beta].sub.6][x.sub.6] + [[beta].sub.7][x.sub.7] + [[beta].sub.11][x.sub.11] + [[beta].sub.12][x.sub.12] + [[beta].sub.14][x.sub.14] + [[beta].sub.16][x.sub.16] + [D.sub.1] + [D.sub.2] + [D.sub.3] + [epsilion]

where pocx is weather, [x.sub.4] employment, [x.sub.5] wages + salaries, [x.sub.6] GD[P.sub.real], [x.sub.7] taxes, [x.sub.11] subsidies, [x.sub.12], coal prices, [x. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.