Question

22. We fit a simple linear regression model using price (in dollars) to predict the number of packets of dog biscuits sold per day. The regression equation is y = 98.1 - 9.8x, and R2 = 0.5275.

Explain how to interpret the R2 in the context of this problem.: *
(A) 52.75% of the variation in the price is explained by the number of packets of dog biscuits sold per day.
(B) 52.75% of the variation in the number of packets of dog biscuits sold per day is explained by the price.
(C) The model is correct 52.75% of the time.
(D) If there is no association between the number of packets of dog biscuits sold and the price, we have a probability of 0.5275 of getting a slope of -9.8, or a more extreme result.

23. Fifty children were selected at random from students at an elementary school. Each of these students was classified according to sugar consumptions (high or low) and exercise level (high or low). The resulting data are summarized in the following frequency table. Using the resulting data summarized in the frequency table below, determine the row conditional relative frequency for a child with high sugar consumption and low exercise level.: *

(A) 14/50 = 0.280
(B) 14/32 = 0.4375
(C) 18/50 = 0.360
(D) 18/32 = 0.5625

24. What is the best method for detecting constant variance in the residuals?: *
(A) Approximate straight line on a QQ-plot
(B) Plot of explanatory variable versus response variable
(C) Plot of fitted values versus residuals
(D) Histogram of the residuals

Answer 1

22) By R^2 we say the variability of y explained by X. So our y is the number of packets of dog biscuits and x is the price. So the answer is B.

(B) 52.75% of the variation in the number of packets of dog biscuits sold per day is explained by the price

25) You forgot to upload the frequency table. So I tell you how to find the answer

the answer is given by

NO. of students with high sugar consumption and low exercise level/50

is the answer

24) we can detect about constant variance by plotting the fitted values versus residual. If we find out some pattern or we can say non randomness then there we say that the errors do not have constant variance.

So the answer is C.

(C) Plot of fitted values versus residuals