Academic Integrity: tutoring, explanations, and feedback — we don’t complete graded work or submit on a student’s behalf.

One strategy to develop and test a statistical model is to divide the data into

ID: 3201931 • Letter: O

Question

One strategy to develop and test a statistical model is to divide the data into a training data a set used to fit the model and a test set used to assess the fitted model's accuracy. In of the following situations name the sampling method used to divide the data from a list ot subjects into a training set and a test set with an allocation of 2/3 and 1/3, respectively. A random number between 1 and 3 is chosen. Starting with that subject, every third subject is assigned to the test set. All others are assigned to the training set. A random number is assigned to each subject on the list using a computer random number generator. Subjects are ordered by their random number, and the first 2/3 are assigned to the training set. All others are assigned to the test set. Subjects assigned random numbers in (b) are divided into two lists by gender. Within each list, the first 2/3 of the subjects ordered by random number are assigned to the training set. All others are assigned to the test set.

Explanation / Answer

a) it is systeematic sampling, As random sample has been chosen sequentially and based on every kth consecutive appearance.

b) it is random sampling as number is chosen by random number given by computer,

c) it is stratified sampling as sample has been divided into strata by their gender charecteristic.