STAT 3006 Lecture Notes - Lecture 19: Homoscedasticity, Standard Deviation, Analysis Of Variance
Document Summary
Statistics are the variables on the left; parameters are the variables on the right: Recall statistics are random variables (so we need to know the distribution). After we get the values of the data, the statistics can be calculated. Conditions for inference: observations are independent, relationship is linear, standard deviation of y, , is the same for all values of x (homoscedasticity or equal variance), the response y varies normally around its mean. We may look for evidence of a significant relationship between variables x and y in the population from which out data were drawn. For that, we can test the hypothesis that the regression slope parameter (beta) is equal to zero. Testing this null hypothesis also allows us to test the hypothesis of no correlation between x and y in the population. We can test the above null hypothesis verses a one- or two-sided alternative.