DATA ANALYSIS FINAL

What will the value of the variable "x" be after the following R command? x <- 5 + 1^2 - 9

-3.00

Suppose that the following contingency table is saved as the variable "snow" in R. Status Likes Doesnt Child 85 36 Adult 69 52 What will the entry of the upper-left cell in the table produced by the following command be? Round your answer to four decimal places. prop.table(snow,1);

0.7025

Determine the value of the variable x after the R code below has completed. x <- 1 while( x < 10 ) { x <- x + 2 }

11

Suppose that the matrix A is defined by: A <- rbind( c(7,1), c(2,8), c(6,3) ); What are the dimensions of A? Select one: a. 1 by 1 b. 7 by 1 c. 3 by 3 d. 2 by 3 e. 3 by 2

3 by 2

Suppose that a k-NN algorithm is attempting to classify a categorical variable which can be either red or blue. The following confusion matrix is produced. Find the accuracy (percent correctly classified) and state it as a percent between 0 and 100 with one decimal place of accuracy. predictions red blue red 40 10 blue 7 20

77.9

Select the correct terms to complete the sentence below. Using Bayes Theorem we can compute __________ where the event and given information have been __________.

Using Bayes Theorem we can compute [conditional probabilities] where the event and given information have been [reversed].

Which of the following will be the value of x after the sequence of R commands below is run? v.1 <- c(6,3,20,9); x <- v.1[-c(1,3)]; Select one: a. 1 3 b. 3 9 c. 6 20 d. 6 e. 2 4

b. 3 9

According to your textbook, which of the following has a sample statistic which is a "great estimator" for the population parameter? Check all that apply. Select one or more: a. median b. range c. mode d. mean e. standard deviation

mean

Which of the following measures of spread were mentioned in your textbook? Check all that apply. Select one or more: a. The variance b. The frequency distribution c. The percentile d. the range e. The average deviation f. the Inter-Quartile Range

the range, the Inter-Quartile Range, The average deviation, The variance

If the frequency distribution for a variable X is given below, compute the probability that X=b. Round your answer to four decimal places. X Frequency a 46 b 39 c 42

0.3071

A bag contains 6 red slips of paper, 10 blue slips of paper, and 10 white slips of paper. You reach in and pull one slip randomly. What is the probability you pull a blue slip of paper? Round your answer to four decimal places.

0.3846

Drag the words below to correctly fill in the blanks in the following statement from your book. Note that not all words will be used. The smoothing methods of forecasting can be summarized nicely in a table such as the one below. (TOP TITLE ROW): Trend (row) / Seasonality(column) | ___________| Additive | Multiplicative (GRAPH UNITS BELOW) None NN NA NM _______ AN AA AM AdditiveDamped DN DA DM __________ MN MA MM For each of these _______________, there exists a version using either additive or multiplicative errors, and therefore in this taxonomy, there are ______________ methods. Any of these methods can be identified by specifying which type of ______________, ______________, and ______________ (or ETS) component they use. OPTIONS time series 36 3 component trend seasonality error 12 24 additive multiplicative none

(TOP TITLE ROW): Trend (row) / Seasonality(column) |[none]| Additive | Multiplicative (GRAPH UNITS BELOW) None NN NA NM [additive] AN AA AM AdditiveDamped DN DA DM [multiplicative] MN MA MM For each of these [twelve], there exists a version using either additive or multiplicative errors, and therefore in this taxonomy, there are [twenty‑four] methods. Any of these methods can be identified by specifying which type of [error], [trend], and [seasonality] (or ETS) component they use.

Suppose that an event A contains 6 outcomes and an event B contains 11 outcomes. If there are 40 outcomes all together in the sample space, and the events A and B are mutually exclusive (have no outcomes in common), what is the probability of their union? That is, what is P(A∪B)? Round your answer to four decimal places.

0.4250

A fair six-sided die is rolled twice, and the sum of the two rolls is noted. What is the probability that sum is even given that the first roll was odd? Round your answer to four decimal places.

0.5

You wish to test the hypothesis that there is a difference between two population means. Arrange the steps below in the order in which they should be conducted to complete this test. OPTIONS: -If this last check is okay, use the student T‑test, otherwise use the Welch T‑test. -Determine if it is okay to assume the samples came from normal populations. -Collect random samples from each of the two populations. -Determine if it is okay to assume the population variances are equal.

1.Collect random samples from each of the two populations. 2.Determine if it is okay to assume the samples came from normal populations. 3.Determine if it is okay to assume the population variances are equal. 4. If this last check is okay, use the student T‑test, otherwise use the Welch T‑test.

Suppose that data is collected on the age and height of tomato plants. A simple linear regression using age (in days) as the independent variable and height (in inches) as the dependent variable produces the following linear equation. Use it to predict the height of a 33 day old tomato plant. Round your answer to two decimal places. y = 0.49x + 1.17

17.34

Find the value of the variable x after the following sequence of R commands is run. v.1 <- 1:3; v.2 <- c(-2,0,0); x <- sum(v.1 + v.2);

4

Match each claim below with the most appropriate hypothesis test. Assume that all assumptions of any hypothesis test are met. A)The claim that two population means are equal. B)The claim that a population mean is less than a certain value. C)The claim that a population mean is more than a certain value. D)The claim that there is no relationship between two categorical variables E)The claim that three or more population means are equal OPTIONS single sample T-test two sample T-test chi-squared test of independence analysis of variance test

A The claim that two population means are equal. → two-sample T test, B The claim that a population mean is less than a certain value. → single-sample T test, C The claim that a population mean is more than a certain value. → single-sample T test, D The claim that there is no relationship between two categorical variables → chi-squared test of independence, E The claim that three or more population means are equal → analysis of variance test

A(n) __________ is the most basic data structures in R. They are essentially a(n) __________ collection of values of the same type. The process of extracting one or more elements is called _________. vector ordered subsetting character numeric array intersecting evaluating

A(n) [vector] is the most basic data structures in R. They are essentially a(n) [ordered] collection of values of the same type. The process of extracting one or more elements is called [subsetting].

Which of the following best describes how a 95% confidence interval for a mean based on a sample of size n should be interpreted? Select one: a. It is the range of values in which there is a 95% chance the true population mean lies. b. It is the range of values that contains the means of 95% of the possible samples of size n taken from the population. c. It is the range of values that will contain the true population mean for 95 out of 100 samples of size n.

It is the range of values that contains the means of 95% of the possible samples of size n taken from the population.

What assumptions must be made in order to conduct a one-way ANOVA test? Check all that apply. Select one or more: a. The variances in the populations are equal b. The distribution of residuals is approximately normal c. All sample sizes are at least 30 d. All observations are independent

All observations are independent, The distribution of residuals is approximately normal, The variances in the populations are equal

Match the following Boolean expressions with the equivalent English statement. -At least one of p or q is true -p and q are both false -p is not true or q is true -Both p and q are true. Choose from boolean expressions: !p && !q p || q p && q !p && q

At least one of p or q is true→ p || q p and q are both false→ !p && !q p is not true or q is true→ !p || q Both p and q are true.→ p && q

Which of the following statements about covariance are true? Check all that apply. Select one or more: a. Covariance can be bigger than one. b. Covariance can be zero. c. Covariance does not depend on the units in which the variables are given. d. Covariance can be negative. e. Covariance does not depend on the order of the variables (x and y).

Covariance can be zero., Covariance can be negative., Covariance can be bigger than one., Covariance does not depend on the order of the variables (x and y).

Drag the most appropriate term from the selection below to complete each of the following sentences. Terms may be used once, more than once, or not at all. In statistical learning, the _______________ of a model refers to the error of the model introduced by attempting to model a complicated real-life relationship with an approximation. The ______________ of a model refers to how sensitive a model is to changes in the data that built the model. To build a ___________ model we break the training data up into equal groups, train the model on data from all but one of the groups, and then test it on the remaining group. We then repeat this process for each combination of groups. The most common technique to detect multicollinearity is to calculate each predictor variable's ____________. cross‑validation variance inflation factor variance residual regularization interaction term bias overfitting

Drag the most appropriate term from the selection below to complete each of the following sentences. Terms may be used once, more than once, or not at all. In statistical learning, the [bias] of a model refers to the error of the model introduced by attempting to model a complicated real-life relationship with an approximation. The [variance] of a model refers to how sensitive a model is to changes in the data that built the model. To build a [cross‑validation] model we break the training data up into equal groups, train the model on data from all but one of the groups, and then test it on the remaining group. We then repeat this process for each combination of groups. The most common technique to detect multicollinearity is to calculate each predictor variable's [variance inflation factor].

Which of the following statements are true about functions in R? Check all that apply. Select one or more: a. Functions can take more than one argument. b. To name a function in R, assign the "function" to a variable just like any other value c. Functions in R must have a type specified ahead of time. d. You can apply a function to all members of a vector using "sapply"

Functions can take more than one argument., You can apply a function to all members of a vector using "sapply", To name a function in R, assign the "function" to a variable just like any other value

Describe at least two things that you learned about dealing with missing data in the reading and/or videos. for multiple imputations section

I learned that data can either be MCAR, MAR, or MNAR, and this directly affects the procedures we should take to deal with missing data. I also learned about multiple imputations which were described as being one of the best ways to deal with missing data and predictions of them.

Which of the following statements does your author make about Kitchen Sink Regression? Check all that apply. Select one or more: a. It will produce highly predictive models. b. You can always maximize your R2 value using this approach. c. It is a good way to use all available inputs in your analysis. d. It is a naive (and awful) approach to use e. It is akin to throwing all of your variables against a wall to see what sticks.

It is a naive (and awful) approach to use., You can always maximize your R2 value using this approach. , It is akin to throwing all of your variables against a wall to see what sticks. Question 2 Partially correct Mark 0.48 out of 1.00 Not flaggedFlag question Question text

Which of the following does your author identify as myths about the bootstrap method? Check all that apply. Select one or more: a. It is only suitable for Bayesian approaches to estimation. b. It is an assumption-free technique for doing inferential statistics. c. It is an ideal tool to use with small samples. d. It requires us to use sampling with replacement. e. It will work with many different types of statistics.

It is an ideal tool to use with small samples., It is an assumption-free technique for doing inferential statistics.

In your own words, describe what is meant by the Out-Of-Bag (OOB) error rate of a random forest.

It is the error rate of random forest, it is called out of bag because random forest tests use "bagging" to collect their data samples. Therefore out of bag is meaning those that are not included in the sample tested from random forest.

Which of the following is a correct interpretation of the p-value of a hypothesis test? Select one: a. It is the probability of observing something at least as extreme as what we did assuming the null hypothesis is true. b. It is a measure of the size of the difference between the null hypothesis and what we observe. c. It is the probability the null hypothesis is true given what we observed.

It is the probability of observing something at least as extreme as what we did assuming the null hypothesis is true.

It the bootstrap method, we take _________ from our sample and do it ___________. We then use these samples to approximate the_______________ for ______________ statistics. a few important statistics medians with replacement a wide range of haphazardly samples means without replacement sampling distribution

It the bootstrap method, we take [samples] from our sample and do it [with replacement]. We then use these samples to approximate the [sampling distribution] for [a wide range of] statistics.

Select the correct word or phrase to complete the sentence below, which describes the relationship between statistical significance and effect size. Not all _______ between means are ________. Just because there is a ________ does not guarantee that the __________ is large enough to make it important.

Not all [tiny differences] between means are [important]. Just because there is a [statistically significant difference] does not guarantee that the [effect size] is large enough to make it important.

Which types of variables does R allow? Check all that apply. Select one or more: a. String Variables b. Boolean Variables c. Numeric Variables

Numeric Variables, Boolean Variables, String Variables

Below is the plot of the decision tree used in the video examples to classify iris species. Use it to classify each of the irises described below by dragging the species to the indicated spot. Species may be used once, more than once, or not at all. look at tree picture Petal Width: 1.4 and Petal-Length: 3.1-- blank Petal Width: 1.4 and Petal-Length: 2.4 -- blank Petal Width: 1.8 and Petal-Length: 2.1 -- blank virginica versicolor setosa

Petal Width: 1.4 and Petal-Length: 3.1-- [versicolor] Petal Width: 1.4 and Petal-Length: 2.4 -- [setosa] Petal Width: 1.8 and Petal-Length: 2.1 -- [setosa]

Which of the following are valid conclusions for a hypothesis test? Check all that apply. Select one or more: a. Fail to Reject The Null Hypothesis b. Prove the Null Hypothesis c. Disprove the Alternative Hypothesis d. Reject the Alternative Hypothesis e. Disprove the Null Hypothesis f. Fail to Reject the Alternative Hypothesis g. Reject the Null Hypothesis h. Prove the Alternative Hypothesis

Reject the Null Hypothesis, Fail to Reject The Null Hypothesis

Your book mentions several broad categories of relationships between two variables. Which of the following were in that list? Select one or more: a. Relationships between two continuous variables b. Relationships between two categorical variables c. Relationships between one categorical and one continuous variable

Relationships between two continuous variables, Relationships between two categorical variables, Relationships between one categorical and one continuous variable

Some problems with messy data can be easily detected in small data sets by ____________ the data. However, mistakes are notoriously ___________. Furthermore, this is untenable for ___________. We must off-load this process to ___________ OPTIONS hiring a flunkey small data sets easy to find large data sets R easy to miss visually auditing

Some problems with messy data can be easily detected in small data sets by [visually auditing] the data. However, mistakes are notoriously [easy to miss]. Furthermore, this is untenable for [large data sets]. We must off-load this process to [R].

Drag each of the accuracy measure names below to the description of that measure. A ________ =(1/v)∑|et|. This is the magnitude of the average absolute error. B________ =(1/v)∑et. This gives an indication of whether the forecasts are on average over- or under-predicting. C ________ =(1/v)∑∣(et/yt)∣×100. This givs a percentage score of how forecasts deviate from the actual values. D ________ =sqrt[(1/v)∑(et^2)]. This measure has the same units as the data series. [Mean Absolute Error (MAE)] [Root Mean Square Error (RMSE)] [Mean Absolute Percentage Error (MAPE)] [Average Error] [Mean Absolute Scaled Error (MASE)]

The correct answer is: Drag each of the accuracy measure names below to the description of that measure. A[Mean Absolute Error (MAE)] =1v∑t=1v|et|. This is the magnitude of the average absolute error. B[Average Error] =1v∑t=1vet. This gives an indication of whether the forecasts are on average over- or under-predicting. C[Mean Absolute Percentage Error (MAPE)] =1v∑t=1v∣∣∣etyt∣∣∣×100. This givs a percentage score of how forecasts deviate from the actual values. D[Root Mean Square Error (RMSE)] =1v∑t=1ve2t‾‾‾‾‾‾‾‾⎷. This measure has the same units as the data series.

n Bayesian analysis, the hyper-parameters that define the posterior distribution are rarely known. Thus we use the MCMC method to determine the probability density of the posterior distribution for each parameter value. For what does the acronym MCMC stand? MEASURE, MONTE, CARLO, CHAIN, MEAN, CREDIBLE, CONFIDENCE, MARKOV (CHOOSE 4)

The correct answer is: In Bayesian analysis, the hyper-parameters that define the posterior distribution are rarely known. Thus we use the MCMC method to determine the probability density of the posterior distribution for each parameter value. For what does the acronym MCMC stand? [Markov] [Chain] [Monte] [Carlo]

Suppose that a sample of size 85 is taken from a population with standard deviation σ= 3.1. Which of the following will be the standard error for the sample mean? Select one: a. 3.1 / 85 b. 3.1 * sqrt(85) c. 3.1 * 85^2 d. 3.1 / sqrt(85) e. 3.1 / 85^2 f. 3.1 * 85 g. 3.1

The correct answer is: 3.1 / sqrt(85)

ndicate which string would be matched by each of the regular expressions below. There will be only one correct string in each case. The regular expression "Zip\\d+" The regular expression "zip.*" The regular expression "[Zz].*\\s.*" The regular expression "^\\d\\d\\d\\d\\d" OPTIONS 99324 Zip99324 zip99324 Zip 99324

The correct answer is: The regular expression "Zip\\d+" → Zip99324, The regular expression "zip.*" → zip99324, The regular expression "[Zz].*\\s.*" → Zip 99324, The regular expression "^\\d\\d\\d\\d\\d" → 99324

Which of the following statements are true about the k-Nearest Neighbors classification algorithm? Check all that apply. Select one or more: a. The classification algorithm does not handle non-binary categorical variables (variables with more than two levels). b. If k=1 is chosen, it is likely that the model will overfit the data c. It is best to use k=1 whenever possible. d. The training and testing steps happen simultaneously. e. It returns the consensus classification of the k nearest neighbors

The correct answers are: It returns the consensus classification of the k nearest neighbors., The training and testing steps happen simultaneously., If k=1 is chosen, it is likely that the model will overfit the data.

For each desirable quality described below, indicate which of the four classification algorithms seen in this chapter exhibits that quality the most: This is an example of ensemble learning that can be extraordinarily accurate and easy to create but does not typically overfit the data. Unfortunately, it is not as transparent as the more basic algorithm on which it is built. Since no particular model is imposed on the training data, this is one of the most flexible and accurate classification learners there are and is very widely used. While this algorithm can only handle binary classification, it creates a model with coefficients which we can use to understand why each observation is classified the way it is. These models are one of the easiest to interpret of all of the popular classifiers out there and once it is created, it can be used to classify observations without a complicated computer algorithm. options Random Forests k-Nearest Neighbors Logistic Regression Decision Trees

The correct answer is: This is an example of ensemble learning that can be extraordinarily accurate and easy to create but does not typically overfit the data. Unfortunately, it is not as transparent as the more basic algorithm on which it is built. → Random Forests, Since no particular model is imposed on the training data, this is one of the most flexible and accurate classification learners there are and is very widely used. → k-Nearest Neighbors, While this algorithm can only handle binary classification, it creates a model with coefficients which we can use to understand why each observation is classified the way it is. → Logistic Regression, These models are one of the easiest to interpret of all of the popular classifiers out there and once it is created, it can be used to classify observations without a complicated computer algorithm. → Decision Trees

Match each of the following descriptions of forecasting methods with the name of the method described. This method forecasts future values using the last observed value for every future value. This method forecasts future values by taking exponentially weighted averages of the previous values and adding both a term for the trend and a third term for smoothing the seasonal component. This method forecasts future values by taking exponentially weighted averages of previous values. This method forecasts future values by taking exponentially weighted averages of the previous values and adding a term for the trend. This method forecasts future values using the mean of the historical data. OPTIONS The Mean Method The Simple Exponential Smoothing Method Holt's Linear Method, The Naive Method The Holt-Winters Forecasting Method

The correct answer is: This method forecasts future values using the last observed value for every future value. → The Naive Method, This method forecasts future values by taking exponentially weighted averages of the previous values and adding both a term for the trend and a third term for smoothing the seasonal component. → The Holt-Winters Forecasting Method, This method forecasts future values by taking exponentially weighted averages of previous values. → The Simple Exponential Smoothing Method, This method forecasts future values by taking exponentially weighted averages of the previous values and adding a term for the trend. → Holt's Linear Method, This method forecasts future values using the mean of the historical data. → The Mean Method

Correlation between values of a time series in neighboring periods is called autocorrelation because: Select one: a. it was discovered by forecasters working in the transportation industry. b. it describes a relationship between the series and itself. c. it involves external data. d. it functions independently of the model selected. e. it is computed automatically.

The correct answer is: it describes a relationship between the series and itself.

Suppose that the linear model for tomato plant heights presented above was developed using the heights of tomato plants with a sample of ages between 10 and 60 days. For which of the following ages would the model produce useful predictions? Check all that apply. Select one or more: a. 1 day b. -10 days c. 100 days d. 50 days e. 20 days

The correct answers are: 20 days, 50 days

Which of the following are advantages of Bayesian analysis over a frequentist approach? Check all that apply. Select one or more: a. Bayesian analysis produces an entire distribution instead of just a confidence interval and/or p-value. b. Bayesian analysis allows us to incrementally adjust our results as new evidence is gathered. c. Bayesian analysis allows us to factor in previous evidence and/or assumptions. d. Bayesian analysis is much more widely accepted.

The correct answers are: Bayesian analysis produces an entire distribution instead of just a confidence interval and/or p-value., Bayesian analysis allows us to factor in previous evidence and/or assumptions., Bayesian analysis allows us to incrementally adjust our results as new evidence is gathered.

Which of the following are true of the standard (Pearson) correlation coefficient? Check all that apply. Select one or more: a. Correlation coefficients can be bigger than one. b. Correlation coefficients do not depend on the order of the variables (x and y). c. Correlation coefficients do not depend on the units of the variables. d. Correlation coefficients can be negative e. If the correlation coefficient of two variables is zero, then the variables are not related.

The correct answers are: Correlation coefficients can be negative, Correlation coefficients do not depend on the units of the variables., Correlation coefficients do not depend on the order of the variables (x and y).

What types of data sources did your reading cover? Check all that apply. Select one or more: a. Data stored in SQL databases b. Data stored in XML format c. Data stored in JSON format d. Data stored on paper in file cabinets Feedback

The correct answers are: Data stored in SQL databases, Data stored in JSON format, Data stored in XML format

Which of the following is an accurate description of the significance (or alpha) level of a hypothesis test? Check all that a apply. Select one or more: a. It is the probability of observing a sample as extreme as our sample if the null hypothesis is true. b. It is the probability of making a Type I error. c. It is the degree of confidence we have in our decision. d. It is the probability that we reject the null hypothesis even though it is true. e. It is the probability of making a Type II error.

The correct answers are: It is the probability that we reject the null hypothesis even though it is true., It is the probability of observing a sample as extreme as our sample if the null hypothesis is true., It is the probability of making a Type I error.

Which of the following statements are true of the logistic regression classification algorithm? Check all that apply. Select one or more: a. Logistic regression results in a linear decision boundary. b. Logistic regression actually produces a probability. c. Logistic regression models the residual error using a Bernoulli (or Binomial) distribution. d. Logistic regression uses the ordinary least-squares method to solve for the coefficients in the model. e. Logistic regression only works with a binary categorical variable

The correct answers are: Logistic regression models the residual error using a Bernoulli (or Binomial) distribution., Logistic regression results in a linear decision boundary., Logistic regression only works with a binary categorical variable., Logistic regression actually produces a probability.

Which of the following did your author indicate he considered "messy data?" Check all that apply. Select one or more: a. Numbers in the same column being in different units b. Extreme outliers c. Inconsistent data coding d. Misspelled names in categorical variables e. Many missing values (NAs) f. Any data that invalidates the assumptions of the model g. Mis-recorded data and data entry mistakes

The correct answers are: Many missing values (NAs), Misspelled names in categorical variables, Inconsistent data coding, Numbers in the same column being in different units, Mis-recorded data and data entry mistakes, Extreme outliers

Suppose that a sample of size 30 is taken from populations with various distributions. For which of these population distributions can we approximate the distribution of the sample means (sampling distribution) with a normal distribution? Check all that apply. Select one or more: a. Chi-Squared Distribution b. Exponential Distribution c. Normal Distribution d. Uniform Distribution e. Beta Distribution

The correct answers are: Normal Distribution, Uniform Distribution, Exponential Distribution, Beta Distribution, Chi-Squared Distribution

Which of the probability formulas are valid for all events A and B? Check all that apply. Select one or more: a. P(A∩B)=P(A)P(B) b. P(A∪B)=P(A)+P(B) c. P(A∪B)=P(A)+P(B)−P(A∩B) d. P(A⎯⎯⎯⎯)=1P(A) where A⎯⎯⎯⎯ is the compliment of A. e. P(A⎯⎯⎯⎯)=1−P(A) where A⎯⎯⎯⎯ is the compliment of A. f. P(A∩B)=P(A)P(B∣A)

The correct answers are: P(A∪B)=P(A)+P(B)−P(A∩B) , P(A⎯⎯⎯⎯)=1−P(A) where A⎯⎯⎯⎯ is the compliment of A. , P(A∩B)=P(A)P(B∣A)

Which of the following statements about the coefficient of determination, R2, are valid? Check all that apply. Select one or more: a. R^2 is a measure of how good of a fit the model is. b. R^2 is between 0 and 1 c. R^2 measures the amount of variation in the dependent variable attributable to change in the independent variable. d. The R^2 value depends on the units used in our variables. e. A high R^2 value means that you do not need to worry about visualizing your data -- your linear model will be a good predictor.

The correct answers are: R2 is a measure of how good of a fit the model is. , R2 measures the amount of variation in the dependent variable attributable to change in the independent variable. , R2 is between 0 and 1

Which of the following assumptions are required to build a linear regression model? Check all that apply. Select one or more: a. The errors have a constant variance. b. That multi-collinearity is at a minimum. c. The effect of the independent variables on the dependent variable are linear and additive. d. The error terms are uncorrelated. e. The residuals are normally distributed with a mean of 0 f. That sample sizes are at least 30.

The correct answers are: The residuals are normally distributed with a mean of 0, The error terms are uncorrelated., The errors have a constant variance., The effect of the independent variables on the dependent variable are linear and additive., That multi-collinearity is at a minimum.

Which of the following are components of time series mentioned in your text. Select one or more: a. Trend b. Maximum c. Scalar d. Linear e. Seasonality f. Cyclic g. Error

The correct answers are: Trend, Seasonality, Error, Cyclic

n which of the following situations does it make sense to use Bayesian methods? Check all that apply. Select one or more: a. You have a medium amount of evidence, and different posteriors illustrate the updated beliefs from a diverse array of prior beliefs. b. You have very little evidence, but you have to make an important decision given the evidence. c. You have a lot of evidence, and it makes no real difference which prior any reasonable person uses, because the evidence will overwhelm it.

The correct answers are: You have a lot of evidence, and it makes no real difference which prior any reasonable person uses, because the evidence will overwhelm it., You have very little evidence, but you have to make an important decision given the evidence. , You have a medium amount of evidence, and different posteriors illustrate the updated beliefs from a diverse array of prior beliefs.

Which of the following are parameters of the family of binomial distributions? Select one or more: a. p -- the probability of a success b. mu -- the mean c. k -- the number of successes d. n -- the number of trials e. sigma -- the standard deviation

The correct answers are: n -- the number of trials, p -- the probability of a success

Which of the following will result in a vector containing the numbers 3, 4, and 5 in that order? Check all that apply. Select one or more: a. c(3,5,1) b. seq(3,5,1) c. c(3,4,5) d. 3:5 e. [3-5]

The correct answers are: seq(3,5,1), 3:5, c(3,4,5)

Which of the following assumptions must be true in order to conduct a one-sample T-test? Check all that apply. Select one or more: a. We must have both np and n(1−p) greater than 5. b. The variance of the population must be known (or the sample size must be large enough). c. The sample must be independent (this basically means random). d. The data must be sampled from a normal distribution (or the sample size must be large enough). e. The variances must be equal.

The data must be sampled from a normal distribution (or the sample size must be large enough)., The sample must be independent (this basically means random).

What is the typical relationship between the mean and median of a dataset that is symmetric? Select one: a. The mean is larger than the median b. The mean is the negative of the median c. There is no systematic relationship d. The mean and median are approximately equal e. The mean is smaller than the median

The mean and median are approximately equal

Determine if each of the following statements is true for probability mass functions (PMFs), also called probability distributions, for probability density functions (PDFs). for both, or for neither. a)The total area is equal to one b)They can take on negative values c)They can take on values bigger than one d)We use an integral to compute probabilities e)The value of the function is the probability of the input Options: PMF, PDF, BOTH, NEITHER

The total area is equal to one → Both, They can take on negative values → Neither, They can take on values bigger than one → PDFs, We use an integral to compute probabilities → PDFs, The value of the function is the probability of the input → PMFs

Match the features of a box plot listed below with the aspect of the data they represent. a)The upper edge of the box b)The upper end of the top whisker c)The horizontal line inside the box d)The lower edge of the box e)A dot above or below the whiskers f)The lower end of the bottom whisker OPTIONS oulier, median, 1st quartile, maximum, minimum, 3rd quartile

The upper edge of the box → 3rd quartile, The upper end of the top whisker → maximum (excluding outliers), The horizontal line inside the box → median, The lower edge of the box → 1st quartile, A dot above or below the whiskers → outliner, The lower end of the bottom whisker → minimum (excluding outliers)

Which of the following does your author suggest is the most legitimate criticism of Bayesian methods? Select one: a. Bayesian methods require that we make assumptions about the distribution in question. b. There is no guarantee that the MCMC method will converge to a solution. c. Using Bayesian methods requires a deep understanding of mathematics and computer science. d. Bayesian methods require high-powered computers to perform the complex computations.

There is no guarantee that the MCMC method will converge to a solution.

Fill in the blanks with the missing words to complete the following paraphrase from your text. We use __________ to describe the behavior of random variables because they make it easy to compute with and give us a lot of information about how a variable behaves. But before we perform computations, we have to specify the __________

We use [probability distributions] to describe the behavior of random variables because they make it easy to compute with and give us a lot of information about how a variable behaves. But before we perform computations, we have to specify the [parameters].

In the first part of chapter 7, your author uses examples with evidence from a binomial distribution and priors with a beta distribution. Why are those two distributions chosen? Select the one answer which best explains this. Select one: a. Because it is easy to multiply the two together and obtain the posterior distribution. b. Becuase the only situations in which a Bayesian analysis is appropriate are those involving these two distributions. c. Because these are your author's two favorite distributions. d. Because the beta distribution can take on any shape we wish based on the selection of hyper parameters.

Your answer is correct. The correct answer is: Because it is easy to multiply the two together and obtain the posterior distribution.

Match each of the variables below with the type of graph which should be used to summarize its distribution. a)Zipcode b)Favorite Pizza Topping c)Student Age d)Number of Siblings e)Test Average Options: Bar Graph Histogram

Zipcode → Bar Graph, Favorite Pizza Topping → Bar Graph, Student Age → Histogram, Number of Siblings → Histogram, Test Average → Histogram

Match the following descriptions with the corresponding term used in probability. a)The set of all possible results that could happen in an experiment b)Sets of results whose occurrences do not affect each others' probability c)One individual result that could happen in an experiment d)A subset of all the results that could happen in an experiment

a) Outcome b)Independent c)event d)Sample Space

Match the subset commands for the matrix A below with the vectors they produce. [,1] [,2] [,3] [1,] 2 7 1 [2,] 0 2 4 [3,] 1 3 5 options: A.matrix[2:3,1:2] A.matrix[2,] A.matrix[1:2,2:3] A.matrix[2,2] A.matrix[,2] --------------------------------------------------- a) [,1] [,2] [1,] 7 1 [2,] 2 4 b) [1] 7 2 3 c) [,1] [,2] [1,] 0 2 [2,] 1 3 d) [1] 0 2 4 e) [1] 2

a)A.matrix[1:2,2:3] b)A.matrix[,2] c) A.matrix[2:3,1:2] d)A.matrix[2,] e)A.matrix[2,2]

Identify the type of each of the following variables. Types may be used one or more times or not at all. a)The color of a ball b)Weight of an airline passenger c)The sum of two dice rolls d)Time spent on this question e)Favorite Breed of Cat f)Number of Siblings g)Phone Number Options: Categorial Continuous Numerical Discrete Numerical

a)Categorial b)Continuous Numerical c)Discrete Numerical d)Continuous Numerical e)Categorial f)Discrete Numerical g)Categorial

If you wish to find out more about a particular R command (for example, the plot command), you can ask for help by typing ____(a)_____. If you don't remember the exact name of the function you're looking for, but you have an idea what the name should be (perhaps it involves creating a plot), you can type _____(b)______

a)[help("plot")] b)[help.search("plot")]

Match each type of relationship with the graph/visualization with which it could be most easily analyzed. a)Relationships between a categorical and a continuous variable b)Relationships between two continuous variables c)Relationships between two categorical variables

a)box plot b)scatter plot c)Mosaic plot

Determine the percent of values in a normal distribution which fall into the ranges described below. a)below the mean b)within 2 standard deviations of the mean c)within 1 standard devation of the mean d)within 3 standard deviations of the mean

below the mean → 50%, within 2 standard deviations of the mean → 95%, within 1 standard devation of the mean → 68%, within 3 standard deviations of the mean → 99.7%

Which of the following describes Simpson's Paradox? Select one: a. The set of all sets both contains itself and does not. b. Only 23 randomly-chosen people are needed to make the probability two of them share the same birthday at least 50%. c. A sphere can be cut into five pieces which can be reassembled into two spheres with volumes equal to the original sphere. d. A trend which appears in different groups but reverses in their combination.

d. A trend which appears in different groups but reverses in their combination.

The contour plot below shows the joint posterior distribution for two parameters in a Bayesian analysis. Drag the text the region with the indicated level of probability density. Background image for dragging markers onto (look at saved image): low medium high

high middle, moderate next, low outskirts