BIOL361 Lecture Notes - Lecture 18: Multivariate Normal Distribution, Analysis Of Variance, Statistical Population
24 Apr 2016
School
Department
Course
Professor
Document Summary
Standard curve prediction use values of x independent to predict value of y with the fitted line by regression. Simple linear regression (slr) response variable is the outcome single predictor variable x. X vs y graph is a straight line. 1 response variable, but 2 predictor variables (x) Simple linear regression linear model with single predictor variable (x) plotted on horizontal cuase. 1 response or dependent variable, plotted on ordinate = the effect. Model 2 regression error in predictor variable is large, exceeding error in response variable appropriate where both variables are estimated using major-axis or a reduced major axis approach. Tested hypothesis: lipid composition of cell membrane affects sensitivity of the muscle for insulin n= 13. Least squares regression line of all the possible lines that can be drawn on the plot, it has the smallest sum of squared vertical distances between each of the data points and line. Y = a + bx a= y intercept ( x=0)
