Livable Cities - London AMPS | City, University of London Page 79 Variable % Variable % Variable % Variable % HH Size Nuclie 10.7 <127 13.15 None 11.5 <300 0.8 Single 64.8 127-253 28.7 Car 27.2 300-500 18.9 Double 16.3 253-380 34.5 Motorbike 43.3 501-1000 45.9 Composite 8.2 381-506 14.7 Bicycle 8 1001-1500 27 Monthly Income >505 8.9 --Other 10 1501-2000 4.9 <20,000 22 >2000 2.5 20000-60000 53 Single 16.8 Yes 20.5 >60 25 Double 52.4 No 36.1 100-300 8 Occupation Triple 30.6 Sometimes 41.8 301-500 35.8 Govt. employee 36.1 501-700 21 Pvt. Employee 9.9 Age of Building Owned Vehicle 49.8 701-900 11 Businessman 25.1 Before 2010 32.9 Metrobus 31 901-1100 15 Unemployed 5 2010-2012 26.2 Other PT 11.1 1101-1300 4 Student 11.2 2013-2015 39.9 Foot 8.1 1301-1500 4 Retired 7.4 2016-2018 1.6 1501-1700 0.2 Other 5.3 2019-2021 0 <5 km 1.3 >1700 1 5-10 km 19 <3 2.5 11-20 km 58 >300 17.5 4 15.1 21-30 km 15.7 300-500 12.3 5 22.5 >30 km 6 501-1000 19 6 35.4 1001-1500 26 7 16.7 Everyday 20.5 1501-2000 9.4 >7 7.8 5 days a week 17.2 2001-2500 11 6 days a week 28.4 >2500 4.8 Cement Concrete 37.1 Only weekends 23.8 Cement Brick 29.9 Few times in a month 8.5 Only Brick 22.3 Only Concrete 3.3 Convenient 23 Other 7.4 Comfortable 8.2 Cost effective 12.3 Cement Concrete 25.6 High speed 18 Cement Brick 19.9 Accessible 11.5 Drywall Plaster 23.4 All of the above 23 Only Brick 29 Other 2.5 Other 2.1 Yes 8.1 No 91.9 Yes 9.8 No 90.2 Aluminium single glazed 48.9 Aluminium double glazed 3.8 Wood Clad 47.3 uPVC 0 5-10 71.8 11-20 27.2 >20 1 5-10 16.3 11-20 51.1 >20 32.6 Building Energy Consumption Monthly Electricity Consumption (kwh Monthly Gas Consumption (kwh) Energy Consump. (gallons to kwh eqv.) No. of electrical appliances Transport Characteristics Vehicle Ownership Use of Metrobus Mode Choice Distance travelled/day Frequency of Travel Mode Choice Reason Building Material Outside Building Material Inside Building Insulation Window Insulation Window Material No. of heating appliances Socio-Demographic Building Characteristics Area (sq. m) No. of Storeys No. of Rooms Table 5. Results of descriptives. Multinomial Regression This study uses multinomial regression to analyze and predict categorical outcomes, effectively identifying factors influencing preferences across categories. Variables such as household size, income, storeys, appliances, insulation quality, building materials, transportation mode, and travel distance were examined for their impact on product choices. The method estimates the likelihood of selecting each category, highlighting the predictors' relative influence. Table 6 presents parameter estimates for operational and transport energy use categories, detailed by household size, income, storeys, insulation, and building materials. Operational Energy Use The baseline intercept values represent the log-odds of a household being in each energy use category (<300, 300-500, 501-1000, 1001-1500, 1501-2000, >2000 kWh) when predictors are at reference levels. Nuclear families are more likely to fall into the 300-500 and 1001-1500 kWh categories, with significant positive coefficients (β = 1.72 and β = 1.92). Single households show strong positive effects in these ranges (β = 2.08 and β = 2.49). In contrast, double households are less likely to be in higher energy categories compared to composite households. Households earning below PKR 20,000 are less likely to fall into higher energy categories, while those earning PKR 20,000-60,000 have a greater likelihood, indicating a positive income-energy use correlation. Single-storey buildings are generally less likely to be in higher energy categories, while double-storey buildings show mixed effects. Households with 5- 10 and 11-20 appliances are more likely to fall into 300-500 and 1001-1500 kWh categories, while 5- 10 heating appliances lower the likelihood of higher energy use, suggesting efficient heating reduces consumption. Insulated buildings consistently show lower probabilities of high energy use, highlighting
