The idea is that more in shape people will have less severe surges in heart rate. This is the change from running to the max pulse during the run. We can then think of the change as a measure of stress on someone's body, accounted for differences in natural resting pulse. This is the change from resting to running pulse. I am going to replace them with two features that I believe will present their meaning more clearly to the model. # Looking for Collinearity of Our PredictorsĪs we see in Plot 1, Max Pulse and Running Pulse are. Nums <- vapply( df, is.numeric, FUN.VALUE = logical( 1))ĭf <- round( df, digits = digits)įitness = read.table( "fitness.dat ", sep = " \t ", header = T)įitness.lm = lm( OXY ~.
We see immediately the model is at least a decent fit to the data, but looking at Tables 1 & 2, we see that most of the significance is driven by the pulse data and age. The data comes from a study conducted at North Carolina State University.įirst, let's glance at the model and see what the fit looks like initially. The danger of overfitting is considerable with only 31 observations, so I am largely assuming there is not interraction unless I see strong evidence otherwise. Given the small size of the dataset, and without strong first principle knowledge of this area, I am not comfortable evaluating interractions. The purpose of this analysis will be to scrutinize the included terms and evaluate potential interractions, using first principle assumptions, as well as diagnostic techniques, to ultimately form a better model than the original. We have a fully fit model with linear additive terms, and no interractions. In this project, I am scrutinizing variables we presume to have an impact on oxygen consumption. Often times, we have an assumption about certain important variables, and we are looking to challenge those assumptions. Title: "Deleting Variables from an Existing Model " Learn more about bidirectional Unicode characters To review, open the file in an editor that reveals hidden Unicode characters. This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below.
0 kommentar(er)
