The TensorTest2D package provides the means to fit generalized linear models on second-order tensor type data. Functions within this package can be used for parameter estimation (e.g., estimating regression coefficients and their standard deviations) and hypothesis testing. We use two examples to illustrate the utility of our package in analyzing data from different disciplines. In the first example, a tensor regression model is used to study the effect of multi-omics predictors on a continuous outcome variable which is associated with drug sensitivity. In the second example, we draw a subset of the MNIST handwritten images and fit to them a logistic tensor regression model. A significance test characterizes the image pattern that tells the difference between two handwritten digits. We also provide a function to visualize the areas as effective classifiers based on a tensor regression model. The visualization tool can also be used together with other variable selection techniques, such as the LASSO, to inform the selection results.
Supplementary materials are available in addition to this article. It can be downloaded at RJ-2022-030.zip
Text and figures are licensed under Creative Commons Attribution CC BY 4.0. The figures that have been reused from other sources don't fall under this license and can be recognized by a note in their caption: "Figure from ...".
For attribution, please cite this work as
Chen, et al., "TensorTest2D: Fitting Generalized Linear Models with Matrix Covariates", The R Journal, 2022
BibTeX citation
@article{RJ-2022-030,
author = {Chen, Ping-Yang and Chang, Hsing-Ming and Chen, Yu-Ting and Tzeng, Jung-Ying and Chang, Sheng-Mao},
title = {TensorTest2D: Fitting Generalized Linear Models with Matrix Covariates},
journal = {The R Journal},
year = {2022},
note = {https://doi.org/10.32614/RJ-2022-030},
doi = {10.32614/RJ-2022-030},
volume = {14},
issue = {2},
issn = {2073-4859},
pages = {152-163}
}