Multivariate adaptive regression splines for analysis of geotechnical engineering systems

Zhang, Wengang; Goh, Alvina

doi:10.1016/j.compgeo.2012.09.016

Cited by 296 publications

(83 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…by applying respective BF(x) with piecewise linear functions: max (0, x -c) where a knot occurs at the position c (Zhang and Goh, 2013). The equation max (.)…”

Section: Multivariate Adaptive Regressions Splinementioning

confidence: 99%

Drought forecasting in eastern Australia using multivariate adaptive regression spline, least square support vector machine and M5Tree model

Deo

Kisi

Singh

2017

Atmospheric Research

254

View full text Add to dashboard Cite

Drought forecasting using standardized metrics of rainfall is a core task in hydrology and water resources management. Standardized Precipitation Index (SPI) is a rainfall-based metric that caters for different time-scales at which the drought occurs, and due to its standardization, is well-suited for forecasting drought at different periods in climatically diverse region. This study advances drought modelling using multivariate adaptive regression splines (MARS), least square support vector machine (LSSVM), and M5Tree models by forecasting SPI in eastern Australia. MARS model incorporated rainfall as mandatory predictor with month (periodicity), SouthernOscillation Index, Pacific Decadal Oscillation Index and Indian Ocean Dipole, ENSO Modoki and Nino 3.0, 3.4 and 4.0 data added gradually. The performance was evaluated with root mean square error (RMSE), mean absolute error (MAE), and coefficient of determination (r 2 ). Best MARS model required different input combinations, where rainfall, sea surface temperature and periodicity were used for all stations, but ENSO Modoki and Pacific Decadal Oscillation indices were not required for Bathurst, Collarenebri and Yamba, and the Southern Oscillation Index was not required for Collarenebri. Inclusion of periodicity increased the r 2 value by 0.5-8.1% and reduced RMSE by 3.0-178.5 %. Comparisons showed that MARS superseded the performance of the other counterparts for three out of five stations with lower MAE by 15.0-73.9% and 7.3-42.2%, respectively. For the other stations, M5Tree was better than MARS/LSSVM with lower MAE by 13.8-13.4% and 25.7-52.2%, respectively, and for Bathurst, LSSVM yielded more accurate result. For droughts identified by SPI  -0.5, accurate forecasts were attained by A C C E P T E D M A N U S C R I P T ACCEPTED MANUSCRIPTMARS/M5Tree for Bathurst, Yamba and Peak Hill, whereas for Collarenebri and Barraba, M5Tree was better than LSSVM/MARS. Seasonal analysis revealed disparate results where MARS/M5Tree was better than LSSVM. The results highlight the importance of periodicity in drought forecasting and also ascertains that model accuracy scales with geographic/seasonal factors due to complexity of drought and its relationship with inputs and data attributes that can affect the evolution of drought events. KeywordsStandardized precipitation index, drought forecasting, multivariate adaptive regression spline, least square support vector machine, M5 Tree model IntroductionDrought is an insidious natural hazard that occurs as a normal, yet a recurrent feature in an arid, semi-arid, desert or rain-forested region (Wilhite et al., 2000a;Keyantash and Dracup, 2002;Vicente-Serrano, 2016). Drought impacts are exacerbated by shifts to warmer and drier conditions, leading to increased water demand compounded by population growth and consequent expansion of industrial, agricultural and energy sectors (McAlpine et al., 2009; IPCC, 2012). As a critical environmental issue, challenges posed by drought elicit increased alertness among hydrologists, a...

show abstract

“…by applying respective BF(x) with piecewise linear functions: max (0, x -c) where a knot occurs at the position c (Zhang and Goh, 2013). The equation max (.)…”

Section: Multivariate Adaptive Regressions Splinementioning

confidence: 99%

Drought forecasting in eastern Australia using multivariate adaptive regression spline, least square support vector machine and M5Tree model

Deo

Kisi

Singh

2017

Atmospheric Research

254

View full text Add to dashboard Cite

show abstract

“…e l s e v i e r . c o m / l o c a t e / e n g g e o Mahdevari and Torabi, 2012;Rafiai and Moosavi, 2012;Zhang and Goh, 2013;Adoko et al, 2013). Though slightly inferior to the MARS and ANN methods in terms of predictive capacity, the regression models remain popular due to their simplicity and model interpretability.…”

Section: Contents Lists Available At Sciencedirectmentioning

confidence: 98%

Regression models for estimating ultimate and serviceability limit states of underground rock caverns

Zhang

Goh

2015

Engineering Geology

Self Cite

View full text Add to dashboard Cite

“…GEP and MARS can overcome the aforementioned deficiencies of the MLR and ANNs approaches and have distinct advantages: 1) they do not act like a black-box, 2) they provide an equation (a functional relation in which the dependent variable is stated directly in terms of the independent variables) between inputs and outputs, 3) they do not need to assume a priori a specific form of function to characterize the physics of the underlying problem, 4) they can capture the complex and nonlinear relationship between a response variable and its predictors and yield accurate results, and finally 5) they are more flexible than the traditional linear and nonlinear regression techniques and usually can overcome their limitations (Johari et al, 2006;Samui et al, 2011, Gandomi andAlavi, 2011;Zhang and Goh, 2013).…”

Section: Introductionmentioning

confidence: 98%

Estimation of soil cation exchange capacity using Genetic Expression Programming (GEP) and Multivariate Adaptive Regression Splines (MARS)

Emamgolizadeh

Bateni

Shahsavani

et al. 2015

Journal of Hydrology

116

View full text Add to dashboard Cite

Multivariate adaptive regression splines for analysis of geotechnical engineering systems

Cited by 296 publications

References 19 publications

Drought forecasting in eastern Australia using multivariate adaptive regression spline, least square support vector machine and M5Tree model

Drought forecasting in eastern Australia using multivariate adaptive regression spline, least square support vector machine and M5Tree model

Regression models for estimating ultimate and serviceability limit states of underground rock caverns

Estimation of soil cation exchange capacity using Genetic Expression Programming (GEP) and Multivariate Adaptive Regression Splines (MARS)

Contact Info

Product

Resources

About