Output similar to ordinary regression for high dimensional data. The method allows testing of non-linear associations alongside feature selection. It also allows testing for "hinges" where associations change in strength and/or direction. It only includes features, for example linear effects of confounders or interactions where model fit is substantially improved. This model fit assessment usually uses cross validation.
Use-case
No response
Is your feature request related to a problem?
No response
Is your feature request related to a JASP module?
Machine Learning, Regression
Describe the solution you would like
implement MARS
Describe alternatives that you have considered
No response
Additional context
Frequentist MARS is implemented in the earth package and the Bayesian counterpart in BASS (with extensions to the original method). I'm happy to talk further about the frequentist implementation, but (apart from knowing of its existence) I know next to nothing of the Bayesian version.
Denison, D. G. T., Mallick, B. K., & Smith, A. F. M. (1998). Bayesian MARS. Statistics and Computing, 8(4), 337–346. https://doi.org/10.1023/A:1008824606259
Francom, D., & Sansó, B. (2020). BASS: An R Package for Fitting and Performing Sensitivity Analysis of Bayesian Adaptive Spline Surfaces. Journal of Statistical Software, 94, 1–36. https://doi.org/10.18637/jss.v094.i08
Francom, D., Sansó, B., Bulaevskaya, V., Lucas, D., & Simpson, M. (2019). Inferring Atmospheric Release Characteristics in a Large Computer Experiment Using Bayesian Adaptive Splines. Journal of the American Statistical Association, 114(528), 1450–1465. https://doi.org/10.1080/01621459.2018.1562933
Francom, D., Sanso, B., Kupresanin, A., & Johannesson, G. (2017). Sensitivity Analysis and Emulation for Functional Data using Bayesian Adaptive Splines. Statistica Sinica. https://doi.org/10.5705/ss.202016.0130
Friedman, J. H. (1991). Multivariate Adaptive Regression Splines. The Annals of Statistics, 19(1), 1–67. https://doi.org/10.1214/aos/1176347963
Friedman, J. H. (1993). Fast MARS (Technical Report 110). Department of Statistics, Stanford University. https://purl.stanford.edu/vr602hr6778
Friedman, J. H., & Roosen, C. B. (1995). An introduction to multivariate adaptive regression splines. Statistical Methods in Medical Research, 4(3), 197–217. https://doi.org/10.1177/096228029500400303
tomtomme
commented
9 months ago
Description
Multivariate adaptive regression splines
Purpose
Output similar to ordinary regression for high dimensional data. The method allows testing of non-linear associations alongside feature selection. It also allows testing for "hinges" where associations change in strength and/or direction. It only includes features, for example linear effects of confounders or interactions where model fit is substantially improved. This model fit assessment usually uses cross validation.
Use-case
No response
Is your feature request related to a problem?
No response
Is your feature request related to a JASP module?
Machine Learning, Regression
Describe the solution you would like
implement MARS
Describe alternatives that you have considered
No response
Additional context
Frequentist MARS is implemented in the earth package and the Bayesian counterpart in BASS (with extensions to the original method). I'm happy to talk further about the frequentist implementation, but (apart from knowing of its existence) I know next to nothing of the Bayesian version.
Denison, D. G. T., Mallick, B. K., & Smith, A. F. M. (1998). Bayesian MARS. Statistics and Computing, 8(4), 337–346. https://doi.org/10.1023/A:1008824606259 Francom, D., & Sansó, B. (2020). BASS: An R Package for Fitting and Performing Sensitivity Analysis of Bayesian Adaptive Spline Surfaces. Journal of Statistical Software, 94, 1–36. https://doi.org/10.18637/jss.v094.i08 Francom, D., Sansó, B., Bulaevskaya, V., Lucas, D., & Simpson, M. (2019). Inferring Atmospheric Release Characteristics in a Large Computer Experiment Using Bayesian Adaptive Splines. Journal of the American Statistical Association, 114(528), 1450–1465. https://doi.org/10.1080/01621459.2018.1562933 Francom, D., Sanso, B., Kupresanin, A., & Johannesson, G. (2017). Sensitivity Analysis and Emulation for Functional Data using Bayesian Adaptive Splines. Statistica Sinica. https://doi.org/10.5705/ss.202016.0130 Friedman, J. H. (1991). Multivariate Adaptive Regression Splines. The Annals of Statistics, 19(1), 1–67. https://doi.org/10.1214/aos/1176347963 Friedman, J. H. (1993). Fast MARS (Technical Report 110). Department of Statistics, Stanford University. https://purl.stanford.edu/vr602hr6778 Friedman, J. H., & Roosen, C. B. (1995). An introduction to multivariate adaptive regression splines. Statistical Methods in Medical Research, 4(3), 197–217. https://doi.org/10.1177/096228029500400303