Table 2 Existing household water consumption models with different sample sizes.
From: Enhancing the explanation of household water consumption through the water-energy nexus concept
Authors | Temporal scale | Technique(s) | Features | Studied period | Studied location | Sample size | R2 |
|---|---|---|---|---|---|---|---|
Bennett et al.20 | Annually | ANN | HI, GEO, WU | 2010 | Queensland, Australia | 205 | 0.30–0.41 |
Jeandorn et al.11 | Daily | Logistic regression | HI, GEO, INF | 2017 | Uvira, Democratic Republic of the Congo | 416 | 0.61 |
Gregory and Leo66 | Annually | SEM | HI, WU | 1996.7–1997.6 | New South Wales, Australia | 471 | 0.33 |
Singha et al.15 | (not reported) | SEM | HI, ATT | 2021 | Fukuoka Prefecture, Japan | 514 | 0.55 |
Lee and Derrible13 | Daily | GBM, OLS, RF, SVM | HI, GEO, WB | 2016 | USA &Canada | 531 | 0.33-0.69 |
Mostafavi et al.9 | Daily | Stepwise regression | HI, WEA, EU, WU | 2009 | USA | 771 | 0.12-0.24 |
Ito et al.14 | Daily | OLS | HI, WEA | 2015.1–2015.4 2015.12–2016.2 2016.8–2016.9 | Kathmandu Valley, Nepal | 992 | 0.26-0.35 |
Duerr et al.21 | Monthly | ARIMA, BART, GBM, RF | GEO, WEA | 1998–2010 | 3 counties in Florida, USA | 973 | – |
Jayarathna et al.8 | Quarterly | OLS | HI, WEA, CAL | 2009–2011 | Queensland, Australia | 1214 | 0.29 |
THIS STUDY | Annually | OLS, RF, XGBoost | HI, WU, EU, EC | 2019 | Beijing, China | 1257 | 0.32-0.52 (Model (4)) |
Bich-Ngoc et al.16 | Annually | OLS | HI, WU | 2014 | Wallonia, Belgium | ~2000 | 0.40–0.42 |
Hoşgör and Fischbeck10 | Annually & Daily | OLS | HI, WEA, CAL | 2009–2011 | Gainesville, USA | 7022 | 0.08–0.14 |
