PMC:2846953 / 25792-26492
Annnotations
{"target":"https://pubannotation.org/docs/sourcedb/PMC/sourceid/2846953","sourcedb":"PMC","sourceid":"2846953","source_url":"https://www.ncbi.nlm.nih.gov/pmc/2846953","text":"MARS is a nonparametric regression model in which no assumption is made regarding the function relationship between dependent and independent variables [22,29]. Instead of residual plots, other approaches are commonly used for assessing nonparametric regression models [29]. Due to the availability of a relatively large sample size in our dataset, we set aside a validation dataset of 20% of the observations and use it to assess the performance of our models. As presented in Table 5 of our manuscript, the R-squared values of the final models based on the analysis dataset are in good agreement with those based on the validation dataset. This approach confirmed the appropriateness of the models.","tracks":[]}