A comparison of random forest regression and multiple linear regression for prediction in neuroscience

Smith, Paul F.; Ganesh, Siva; Liu, Ping

doi:10.1016/j.jneumeth.2013.08.024

Cited by 202 publications

(91 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In addition, MLR models are easy to apply using GIS platforms to quickly produce maps, whereas implementation of RF models can be time-intensive for large geographical areas. The better performances of MLR than RF are also observed in other studies (e.g., [78,79]). …”

Section: Discussionsupporting

confidence: 82%

“…An estimate of error rate in the RF algorithm is obtained from the difference of out-of-bag predictions and corresponding observations. Since RF predictions are out-of-bag cross validated, the variance of predictions is generally larger in RF compared to MLR [78]. This implies that pseudo-R 2 in RF would be less than the R 2 in MLR, when the same set of predictors is used in both approaches.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Evaluating Site-Specific and Generic Spatial Models of Aboveground Forest Biomass Based on Landsat Time-Series and LiDAR Strip Samples in the Eastern USA

et al. 2017

View full text Add to dashboard Cite

Large-area assessment of aboveground tree biomass (AGB) to inform regional or national forest monitoring programs can be efficiently carried out by combining remotely sensed data and field sample measurements through a generic statistical model, in contrast to site-specific models. We integrated forest inventory plot data with spatial predictors from Landsat time-series imagery and LiDAR strip samples at four sites across the eastern USA-Minnesota (MN), Maine (ME), Pennsylvania-New Jersey (PANJ) and South Carolina (SC)-in statistical modeling frameworks to analyze the performance of generic (all sites combined) versus site-specific models. The major objective was to evaluate the prediction accuracy of generic and site-specific models when applied to particular sites. Pixel-level polynomial model fitting was applied to the time-series of near-anniversary date Landsat variables to obtain projected metrics in the target year 2014 for which LiDAR strip samples were available. Two forms of models based on ordinary least-squares multiple linear regressions (MLR) and the random forest (RF) machine learning approach were developed for each site and for the pooled (i.e., generic) reference data frame. The models were evaluated using national forest inventory (NFI) data for the USA. We observed stronger fit statistics with the MLR than with RF for both the site-specific and the generic models. The proportions of variances explained (adjusted R 2 ) with the site-specific models were 0.86, 0.78, 0.82 and 0.92 for ME, MN, PANJ and SC, respectively while the generic model had adjusted R 2 = 0.85. A test of statistical equivalence of observed and predicted AGB for the NFI locations did not reveal equivalence with any of the models, possibly due to the different resolutions of the observed and predicted data. In contrast, predictions by the generic and site-specific models were equivalent. We conclude that a generic model provides accuracies comparable to the site-specific models for large-area AGB assessment across our study sites in the eastern USA.

show abstract

Section: Discussionsupporting

confidence: 82%

Section: Discussionmentioning

confidence: 99%

Evaluating Site-Specific and Generic Spatial Models of Aboveground Forest Biomass Based on Landsat Time-Series and LiDAR Strip Samples in the Eastern USA

et al. 2017

View full text Add to dashboard Cite

show abstract

“…We found that RF performs better than multiple regression at explaining metrics describing forest patch patterns (PD and ENN) and broader landscape structures (GRD and CA). Given the well-established advantages of decision-tree-based methods over those of classical multiple regression (Breiman et al 1984;Breiman 2001;Prasad et al 2006;Cutler et al 2007Cutler et al , 2008Pitcher et al 2011;Ellis et al 2012b;Cutler 2013;Smith et al 2013), we suggest that the reasons for these differences are likely to be because the patch-pattern metrics and broader landscape structures vary in less smooth or monotonic ways (McGarigal et al 2016)-ways that RF is able to capture, but multiple regression is not. Accordingly, we have shown that RF provides a promising methodology for identifying these relationships, and that it has the potential to be an effective tool for providing essential information for aiding land use management decisions, not only in terms of planning, but also for conservation actions, as proposed by Zanella et al (2012), in cases of high rates of anthropogenic biodiversity loss, as it is the case of the Atlantic Forest.…”

Section: Resultsmentioning

confidence: 88%

How well does random forest analysis model deforestation and forest fragmentation in the Brazilian Atlantic forest?

et al. 2017

View full text Add to dashboard Cite

We assessed the value of applying random forest analysis (RF) to relating metrics of deforestation (DF) and forest fragmentation (FF) to socioeconomic (SE) and biogeophysical (BGP) factors, in the Brazilian Atlantic Forest of Minas Gerais, Brazil. A vegetation-monitoring project provided land cover maps, from which we derived DF and FF metrics. An ecologic-economical zoning project provided more than 300 SE and BGP factors. We used RF to identify relationships between these sets of variables and compared its performance in this task to that of a more traditional multiple linear regression approach. We found that RF modelled relatively-well variance in all metrics used (the rate of deforestation, the amount of forest and the density and isolation of forest patches), presenting a better performance when compared to the classical approach. RF also identified geographical location and topographic factors as being most closely associated with patterns of DF and FF metrics. Both analyses found factors associated with economic productivity, social institutions, accessibility and exploration to have little relationship with forest metrics. RF was better at explaining variations in rates of deforestation, remaining forest and patch patterns, than the multiple linear regression approach. We conclude that RF provides a promisHandling Editor: Pierre Dutilleul. Electronic supplementary materialThe online version of this article (https://doi.org/10.1007/s10651-017-0389-8) contains supplementary material, which is available to authorized users.

show abstract

“…The objective of decision tree is to find the interaction between variables, and the weakness of the neural network is its inability to explain its reasoning process and reasoning basis 18 . Compared with the traditional intelligence algorithms, such as ANN and SVM, RFR has high prediction accuracy and good tolerance to outliers and noises 18,19 . Because of its superior performance, RFR has been widely applied to various fields such as biology, medicine, economics, management, remote sensing and other fields in recent years 19–25 .…”

Section: Introductionmentioning

confidence: 99%

An approach to predict the height of fractured water-conducting zone of coal roof strata using random forest regression

Zhao

2018

Sci Rep

View full text Add to dashboard Cite

Water inrushes from coal-roof strata account for a great proportion of coal mine accidents, and the height of fractured water-conducting zone (FWCZ) is of significant importance for the safe production of coal mines. A novel and promising model for predicting the height of FWCZ was proposed based on random forest regression (RFR), which is a powerful intelligent machine learning algorithm. RFR has high prediction accuracy and is robust in dealing with the complicated and non-linear problems. Also, it can evaluate the importance of the variables. In this study, the proposed model was applied to Hongliu Coal Mine in Northwest China. 85 field measured samples were collected in total, with 60 samples (70%) used for training and 20 (30%) used for validation. For comparison, a support vector machine (SVM) model was also constructed for the prediction. The results show that the two models are in accordance with the field measured data, and RFR shows a better performance on good tolerance to outliers and noises and efficiently on high-dimensional data sets. It is demonstrated that RFR is more practicable and accurate to predict the height of FWCZ. The achievements will be helpful in preventing and controlling the water inrushes from coal-roof strata, and also can be extended to various engineering applications.

show abstract

A comparison of random forest regression and multiple linear regression for prediction in neuroscience

Cited by 202 publications

References 15 publications

Evaluating Site-Specific and Generic Spatial Models of Aboveground Forest Biomass Based on Landsat Time-Series and LiDAR Strip Samples in the Eastern USA

Evaluating Site-Specific and Generic Spatial Models of Aboveground Forest Biomass Based on Landsat Time-Series and LiDAR Strip Samples in the Eastern USA

How well does random forest analysis model deforestation and forest fragmentation in the Brazilian Atlantic forest?

An approach to predict the height of fractured water-conducting zone of coal roof strata using random forest regression

Contact Info

Product

Resources

About