Consider a binary response variable y and two predictor variables x1 and x2. The following table contains the parameter estimates of the linear probability model (LPM) and the logistic regression model, with the associated p-values shown in parentheses. Variable LPM Logistic Intercept −0.50 −1.90 (0.02) (0.06) x1 0.26 0.99 (0.06) (0.06) x2 −0.06 −0.30 (0.01) (0.04)   What is the predicted probability implied by the linear probability model for x1 = 8 with x2 equal to 10 and 25?                                       y x1 = 8, x2 = 10       _______ x1 = 8, 2 = 25         _______

x 5.7 4.1 6.2 4.4 6.5 5.8 4.9 y 1.9 4.8 0.8 3.9 1.2 1.7 3.0 (a) Computethecoefficientofdetermination. (b) Howmuchofthevariationintheoutcomevariableisexplainedbytheleast-squares regression line?

Which of the following is(are) TRUE if the coefficient of determination between the response and the predictor variables is 0.81 based on a random sample of size n. O Neither Anor B O A. This means that 19% of the total variation in the response variable remains unexplained by the simple linear regression model that uses the given predictor variable. O Both A and B B. The Pearson's sample correlation coefficient is +0.9.

Suppose x1 ans x2 are predictor variables for a response variable y. a. The distribution of all possible values of the response variable corresponding to particular values of the two predictor variables is called a distribution of the response variable. b. State the four assumptions for multiple linear regression inferences.

Parts b, c, and d only please.

In a study investigating maternal risk factors for congenital syphilis, syphilis is treated as a binary outcome variable, where 1 represents the presence of disease in a newborn and 0 represents absence of disease. The estimated coefficients from a logistic regression model containing the predictors cocaine or crack use, marital status, number of prenatal visits to a doctor, alcohol use and level of education are included in the table below. The estimated intercept is not included in the table. a. As an expectant mother’s number of prenatal visits to the doctor increases, does the probability that her child will be born with congenital syphilis increase or decrease?b. Marital status is a binary variable, where 1 indicates that a woman is unmarried and 0 indicates that she is married. What are the estimated relative odds that a newborn will suffer from syphilis for unmarried versus married mothers after holding the other variables in the model constant?c. Cocaine or crack use is also a…

The following table gives the sales of CD_Rom drives and compact disk players for selected years and can be found in Glassman. Sales are in millions. Year 1988 1990 1992 1993 Sales 0.075 0.24 1.5 4.8 A. On the basis of this data, find the best-fitting exponential regression. Let x = 0 correspond to 1988. State the square of the correlation coefficient. (Use 4 decimal places in you answers.)S(x) =        r2 =        B. Use this model the estimate the sales in 1997. (Round to the nearest player.)In 1997, there were about         million CD_ROM and compact disk players sold.

Assume a multiple linear regression y = Bo + B1 a1+ B2x2 + e. Which statement(s) is(are) true about the variance inflation factors (VIFS) of the coefficient estimates b1 and b2 ? I. The VIF of b, is the same as the VIF of b2. II. VIF will likely be large if X2 is highly positively correlated with X1 II. VIF will likely be large if X2 is highly negatively correlated with X1 IV. VIF will likely be close to 1 if X1 and X2 are independent O l and IV 1, II, III and IV Il and III OIV only I only

2. This question is about logistic regression. The outcome variable is success (1=success, 0=failure), and we have two groups of observations: group A and group B. a. Assume that on average, the odds of success is x for group A members. What is the average probability of success (p) for group A? Please write down p as a function of x. Given an x, is there a unique p? b. Assume that on average, the odds ratio for success of group A over group B is 2. The average probability of success for group A is pA and that the average probability of success for group B is pg. Is the set of Paand pg unique? c. The logistic regression of success ~ group (link=logit) returns a coefficient estimate of -1.3863 for the dummy variable group (1=group A, 0=group B). Interpret this coefficient estimate.