HGHEAT contains the heating costs of tenants. It is collected since Wave C. Until Wave BD, only tenants who stated to pay a rent were asked. Since Wave BE, every tenant was asked. Heating costs are reported in Euro for all years.
In Wave BF, HGHEAT has not been asked for directly. Instead, respondents stated their costs for different energy sources (district heating, gas, liquefied gas, electricity, heating oil, coal, wood, biomass, solar, and geothermal heat) and whether they use them for heating, warm water, cooking, and/or electronic devices respectively (if applicable). To generate HGHEAT, costs for cooking and electricity costs other than heating have been excluded by regressing the usage-dummies and controlling for household-specific (number of people living in the household, type of tenant) and dwelling-specific variables (number of rooms, size of the dwelling, construction year, living environment). The shares for heating and warm water were then added to generate HGHEAT. Due to changes in the questionaire design, firstly, HGHEAT is lower in Wave BF than it has been in previous and following years and secondly, HGHEAT from Wave BF is neither carried backward nor forward.
In file HGEN, only HGHEAT is distributed. All energy cost variables can be found in the dataset BFH.
Missing values of HGHEAT are treated as follows: In a first step, past values of the two last years are carried forward and adjusted for inflation given that the household still lives at the same address, the dwelling's size has remained the same and the rent has not changed considerably. In a second step, if HGHEAT is still missing, values of the two subsequent years are carried backwards in the same manner. In a third step, the remaining missing values of HGHEAT are imputed by Stata's chained imputation procedure. Imputed values are not carried forward. In contrast to previous versions of the SOEP, we do not impute HGHEAT for Wave A and Wave B anymore, since data release v32. For more information on the imputation see the see the description of HGRENT.
HGHEAT contains the heating costs of tenants. It is collected since Wave C. Until Wave BD, only tenants who stated to pay a rent were asked. Since Wave BE, every tenant was asked. Heating costs are reported in Euro for all years.
In Wave BF, HGHEAT has not been asked for directly. Instead, respondents stated their costs for different energy sources (district heating, gas, liquefied gas, electricity, heating oil, coal, wood, biomass, solar, and geothermal heat) and whether they use them for heating, warm water, cooking, and/or electronic devices respectively (if applicable). To generate HGHEAT, costs for cooking and electricity costs other than heating have been excluded by regressing the usage-dummies and controlling for household-specific (number of people living in the household, type of tenant) and dwelling-specific variables (number of rooms, size of the dwelling, construction year, living environment). The shares for heating and warm water were then added to generate HGHEAT. Due to changes in the questionaire design, firstly, HGHEAT is lower in Wave BF than it has been in previous and following years and secondly, HGHEAT from Wave BF is neither carried backward nor forward.
In file HGEN, only HGHEAT is distributed. All energy cost variables can be found in the dataset BFH.
Missing values of HGHEAT are treated as follows: In a first step, past values of the two last years are carried forward and adjusted for inflation given that the household still lives at the same address, the dwelling's size has remained the same and the rent has not changed considerably. In a second step, if HGHEAT is still missing, values of the two subsequent years are carried backwards in the same manner. In a third step, the remaining missing values of HGHEAT are imputed by Stata's chained imputation procedure. Imputed values are not carried forward. In contrast to previous versions of the SOEP, we do not impute HGHEAT for Wave A and Wave B anymore, since data release v32. For more information on the imputation see the see the description of HGRENT.