1. Which of the following is characteristic of a parametric model? a. Complexity
ID: 3585352 • Letter: 1
Question
1. Which of the following is characteristic of a parametric model?
a. Complexity of the model grows as the amount of data increases.
b. Given a set of parameters, future predictions are independent of observed data.
c. Model does not make assumptions about the shape or form of the probability distribution from which data are drawn.
d. Predicted results are often difficult to interpret.
2. Tourists in a country’s hotels have stayed an average of 3.5 nights. A tourism analyst wants to test if recent tourist stays have changed from this average. The analyst obtained a random sample of number of nights in hotels recently spent by tourists. Which test should s/he use?
a. Two sample t-test.
b. t-test comparing two population proportions p1 and p2 using a z-statistic.
c. Hypothesis test estimating a population mean , using a t-statistic.
d. Independent sample t-test; groups have unequal variance.
3. A regression model would be applicable for which problem ?
a. Predict the number of purchases your customers will buy of each product.
b. Predict if a product will have more than 10 purchases .
c. Decide if a product is similar to another product.
d. All of the above.
4. What data characteristics are desirable in model inputs?
a. Fields with the same values
b. Fields with a constant value with respect to other values in other fields
c. Fields with the same granularity as cases that want to predict.
d. Fields with unique values.
5. What are two reasons why a logistic regression model would not be an appropriate solution for a problem?
1. __________________________________
2. __________________________________
6. A researcher wants to know whether the recovery period of a patient after knee surgery will be shorter if a patient goes to physical therapy three times a week instead of two times per week. The average recovery time for patients with knee surgery is 7.5 weeks.
What is H0:
What is Ha:
7. The reducible error due to variance is the difference between the expected (or average) prediction of our model and the correct value which we are trying to predict.
T F
8. Unsupervised data mining techniques specify a target variable.
T F
9. The output of an classification data task is continuous.
T F
10. Directed mining models assume that the set of input variables and training records contain the patterns of what the model should predict.
T F
Explanation / Answer
1) option a
Complexity of the model grows as the amount of data increases.
2) option c
Hypothesis test estimating a population mean , using a t-statistic.Because we need to estimate the test based on average nights stayed.
3) option d
All the above statements are applicable to satisfy the regression model.
4) option c
Fields with the same granularity as cases that want to predict, based on model inputs.
5)
1) Because of binary dependent variable conditional distribution y|x is a Bernoulli distribution instid of Gaussian distribution.
2) probabilities of predicted values are restricted to (0,1). Logistic regression predicts the probability of particular outcomes, when compared to logistic and linear regression.
7) True
8) False
For unsupervised data mining technique no need to specify target.The target has to be set for supervised data mining technique.
9) True
output of an classification data task is continuous.
10) True
Related Questions
Navigate
Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.