CFA

MULTIPLE REGRESSION
1
1. Consider the following analysis of variance table:
Source Sum of Squares Df Mean Square
Regression 20 1 20
Error 80 20 4
Total 100 21
The F-statistic for a test of joint significance of all the slope coefficients is closest to:
(A) 5.
(B) 0.2.
(C) 0.05.

2. One of the underlying assumptions of a multiple regression is that the variance of the
residuals is constant for various levels of the independent variables. This quality is
referred to as:
(A) a linear relationship.
(B) homoskedasticity.
(C) a normal distribution.

3. When constructing a regression model to predict portfolio returns, an analyst runs a

regression for the past five year period. After examining the results, she determines that
an increase in interest rates two years ago had a significant impact on portfolio results
for the time of the increase until the present. By performing a regression over two
separate time periods, the analyst would be attempting to prevent which type of
misspecification?
(A) Inappropriate variable form.
(B) Inappropriate variable scaling.
(C) Incorrectly pooling data.

4. A fund has changed managers twice during the past 10 years. An analyst wishes to
measure whether either of the changes in managers has had an impact on performance.
R is the return on the fund, and M is the return on a market index. Which of the
following regression equations can appropriately measure the desired impacts?

Quantitative Methods 1 Multiple Regression

CFA
(A) R = a + bM + c1 D1 + c2D2 + c3D3 + , where D1 = 1 if the return is from the first
manager, and D2 = 1 if the return is from the second manager, and D3 = 1 is the
return is from the third manager.
(B) The desired impact cannot be measured.
(C) R = a + bM + c1D1 + c2D2 + , where D1 = 1 if the return is from the first
manager, and D2 = 1 if the return is from the third manager.

5. An analyst regresses the return of a S&P 500 index fund against the S&P 500, and also
regresses the return of an active manager against the S&P 500. The analyst uses the
last five years of data in both regressions. Without making any other assumptions,
which of the following is most accurate? The index fund:
(A) should have a higher coefficient on the independent variable.
(B) regression should have higher sum of squares regression as a ratio to the total
sum of squares.
(C) should have a lower coefficient of determination.

6. An analyst runs a regression of portfolio returns on three independent variables. These

independent variables are price-to-sales (P/S), price-to-cash flow (P/CF), and price-to-book
(P/B). The analyst discovers that the p-values for each independent variable are relatively
high. However, the F-test has a very small p-value. The analyst is puzzled and tries to figure
out how the F-test can be statistically significant when the individual independent variables
are not significant. What violation of regression analysis has occurred?
(A) serial correlation.
(B) conditional heteroskedasticity.
(C) multicollinearity.

7. What is the expected salary (in $1,000) of a woman with 16 years of education and 10
years of experience?
(A) 65.48.
(B) 59.18.
(C) 54.98.

8. If the return on the industry index is 4%, the stock's expected return would be:
(A) 11.2%.
(B) 9.7%.
(C) 7.6%.

9. The percentage of the variation in the stock return explained by the variation in the
industry index return is closest to:
(A) 63.2%.
(B) 72.1%.
(C) 84.9%.
Quantitative Methods 2 Multiple Regression
CFA
10. Regarding Sophie's statement on multiple regression:
(A) both statements are correct.
(B) only Statement 1 is correct.
(C) only Statement 2 is correct.

11. Based on the credit spread model, if an issuer gets included in the CDX index and
assuming everything else the same, which of the following statements most accurately
describes the model's forecast?
(A) The credit spread on the firm's issue would decrease by 10 bps.
(B) The credit spread on the firm's issue will decrease by 32 bps.
(C) The credit spread on the firm's issue will increase by 32 bps.

12. Which of the following is least likely an assumption of multiple linear regression?
(A) There is no linear relationship between the independent variables.
(B) The dependent variable is not serially correlated.
(C) The error term is normally distributed.

13. Which assumption of multiple regression is most likely evaluated using a QQ plot?
(A) Serial correlation of residuals.
(B) Error term is normally distributed.
(C) Conditional heteroskedasticity.

14. The predicted price of a house that has 2,000 square feet of space and 4 bedrooms is
closest to:
(A) $114,000.
(B) $256,000.
(C) $185,000.

15. The conclusion from the hypothesis test of H0: b1 = b2 = 0, is that the null hypothesis
should:
(A) not be rejected as the calculated F of 40.73 is greater than the critical value of 3.29.
(B) be rejected as the calculated F of 40.73 is greater than the critical value of 3.33.
(C) be rejected as the calculated F of 40.73 is greater than the critical value of 3.29.

16. Which of the following is most likely to present a problem in using this regression for
forecasting?
(A) Heteroskedasticity.
(B) Multicollinearity.
(C) Autocorrelation.

Quantitative Methods 3 Multiple Regression

CFA
17. Which of the following is the most accurate interpretation of the slope coefficient for
size? ARAR:
(A) and index will change by 1.1% when the natural logarithm of assets under
management changes by 1.0.
(B) will change by 0.6% when the natural logarithm of assets under management
changes by 1.0, holding index constant.
(C) will change by 1.0% when the natural logarithm of assets under management
changes by 0.6, holding index constant.

18. Which of the following is the estimated standard error of the regression coefficient for
index?
(A) 2.31.
(B) 0.52.
(C) 1.91.

19. Which of the following is the t-statistic for size?

(A) 0.70.
(B) 3.33.
(C) 0.30

20. Which of the following is the estimated intercept for the regression?
(A) –2.86.
(B) –9.45.
(C) –0.11.

21. Which of the following statements is most accurate regarding the significance of the
regression parameters at a 5% level of significance?
(A) The parameter estimates for the intercept are significantly different than zero. The
slope coefficients for index and size are not significant.
(B) All of the parameter estimates are significantly different than zero at the 5% level
of significance.
(C) The parameter estimates for the intercept and the independent variable size are
significantly different than zero. The coefficient for index is not significant.

22. Which of the following is NOT a required assumption for multiple linear regression?
(A) The error term is normally distributed.
(B) The expected value of the error term is zero.
(C) The error term is linearly related to the dependent variable.

Quantitative Methods 4 Multiple Regression

CFA
23. Consider the following estimated regression equation, with calculated t-statistics of the
estimates as indicated.
AUTOt = 10.0 + 1.25 Plt + 1.0 TEENt – 2.0 INSt
With a Pl calculated t-statistic of 0.45, a TEEN calculated t-statistics of 2.2, and an INS
calculated t-statistic of 0.63.
The equation was estimated over 40 companies, Using 5% level of significance, which
of the independent variables significantly different from zero?
(A) Pl and INS only.
(B) TEEN only.
(C) Pl only.

24. Suppose the analyst wants to add a dummy variable for whether a person has a
business college degree and an engineering degree. What is the CORRECT
representation if a person has both degrees?
Business Engineering
Degree Degree
Dummy Dummy
Variable Variable
(A) 1 1
(B) 0 1
(C) 0 0

25. Which of the following statements regarding the R2 is least accurate?

(A) The adjusted-R2 not appropriate to use in simple regression.
(B) The adjusted-R2 is greater than the R2 in multiple regression.
(C) It is possible for the adjusted-R2 to decline as more variables are added to the
multiple regression.

26. Which of the following is a potential remedy for multicollinearity?

(A) Add dummy variables to the regression.
(B) Omit one or more of the collinear variables.
(C) Take first differences of the dependent variable.

27. Salve runs a regression using the squared residuals from the model using the original
dependent variables. The coefficient of determination of this model is 6%. Which of the
following is the most appropriate conclusion at a 5% level of significance?
(A) Because the test statistic of 7.20 is lower than the critical value of 7.81, we fail to
reject the null hypothesis of no conditional heteroskedasticity in residuals.
(B) Because the test statistic of 7.20 is higher than the critical value of 3.84, we reject
the null hypothesis of no conditional heteroskedasticity in residuals.
(C) Because the test statistic of 3.60 is lower than the critical value of 3.84, we reject
the null hypothesis of no conditional heteroskedasticity in residuals.

Quantitative Methods 5 Multiple Regression

CFA
28. Which of the following misspecifications is most likely to cause serial correlation in
residuals?
(A) Improper variable scaling.
(B) Improper variable form.
(C) Data improperly pooled.

29. Should Salve be concerned about residual serial correlation?

(A) Yes, for two lags only.
(B) No.
(C) Yes, for one lag only.

30. Should Salve be concerned about residual multicollinearity?

(A) Yes, and Salve should exclude either variable SMB or HML from the model.
(B) Yes, and Salve should exclude variable Rm-Rf from the model.
(C) No.

31. Which of the following conditions will least likely affect the statistical inference about
regression parameters by itself?
(A) Multicollinearity.
(B) Model misspecification.
(C) Unconditional heteroskedasticity.

32. One of the main assumptions of a multiple regression model is that the variance of the
residuals is constant across all observations in the sample. A violation of the
assumption is most likely to be described as:
(A) unstable remnant deviation.
(B) heteroskedasticity.
(C) positive serial correlation.

33. Assume that in a particular multiple regression model, it is determined that the error
terms are uncorrelated with each other. Which of the following statements is most
accurate?
(A) Serial correlation may be present in this multiple regression model, and can be
confirmed only through a Durbin-Watson test.
(B) This model is in accordance with the basic assumptions of multiple regression
analysis because the errors are not serially correlated.
(C) Unconditional heteroskedasticity present in this model should not pose a problem,
but can be corrected by using robust standard errors.

Quantitative Methods 6 Multiple Regression

CFA
34. Sutter has detected the presence of conditional heteroskedasticity in Smith's report. This
is evidence that:
(A) the error terms are correlated with each other.
(B) the variance of the error term is correlated with the values of the independent
variables.
(C) two or more of the independent variables are highly correlated with each other.

35. Suppose there is evidence that the variance of the error term is correlated with the
values of the independent variables. The most likely effect on the statistical inferences
Smith can make from the regressions results using financial data is to commit a:
(A) Type I error by incorrectly failing to reject the null hypothesis that the regression
parameters are equal to zero.
(B) Type II error by incorrectly failing to reject the null hypothesis that the regression
parameters are equal to zero.
(C) Type I error by incorrectly rejecting the null hypotheses that the regression
parameters are equal to zero.

36. Which of the following is most likely to indicate that two or more of the independent
variables or linear combinations of independent variables, may be highly correlated with
each other? Unless otherwise noted, significant and insignificantly mean significantly
different from zero and not significantly different from zero, respectively.
(A) The R2 is low, the F-statistic is insignificant and the Durbin-Watson statistic is
significant.
(B) The R2 is high, the F-statistic is significant and the t-statistics on the individual
slope coefficients are insignificant.
(C) The R2 is high, the F-statistic is significant and the t-statistics on the individual
slope coefficients are significant.

37. Using the Durbin-Watson test statistic, Smith rejects the null hypothesis suggested by
the test. This is evidence that:
(A) two or more of the independent variables are highly correlated with each other.
(B) the error term is normally distributed.
(C) the error terms are correlated with each other.

38. Which model would be a better choice for making a forecast?

(A) Model ONE because it has a higher R2.
(B) Model TWO because it has a higher adjusted R2.
(C) Model TWO because serial correlation is not a problem.
Quantitative Methods 7 Multiple Regression
CFA
39. Using Model ONE, what is the sales forecast for the second quarter of the next year?
(A) $56.02 million.
(B) $51.09 million.
(C) $46.31 million.

40. Which model misspecification is most likely to cause multicollinearity?

(A) Inappropriate variable form.
(B) Ommission of important variable(s).
(C) Inappropriate variable scaling.

41. If it is determined that conditional heteroskedasticity is present in model one, which of

the following inferences are most accurate?
(A) Both the regression coefficients and the standard errors will be biased.
(B) Regression coefficients will be unbiased but standard errors will be biased.
(C) Regression coefficients will be biased but standard errors will be unbiased.

42. Mercado probably did not include a fourth dummy variable Q4, which would have had
0, 0, 0, 1 as its first four observations because:
(A) it would not have been significant.
(B) the intercept is essentially the dummy for the fourth quarter.
(C) it would have lowered the explanatory power of the equation.

43. If Mercado determines that Model TWO is the appropriate specification, then he is
essentially saying that for each year, value of sales from quarter three to four is
expected to:
(A) grow by more than $1,000,000.
(B) remain approximately the same.
(C) grow, but by less than $1,000,000.

44. The adjusted R2 of Model 2 is closest to:

(A) 0.36.
(B) 0.37.
(C) 0.39.

45. The model better suited for prediction is:

(A) Model 1 because it has a lower Bayesian information criterion.
(B) Model 2 because it has a lower Akaike information criterion.
(C) Model 2 because it has a higher Akaike information criterion.

Quantitative Methods 8 Multiple Regression

CFA
46. The F-statistic for testing H0: coefficient of LIQ = 0 versus Ha: coefficient of LIQ # 0 is
closest to:
(A) 5.45.
(B) 13.33.
(C) 2.11.

47. What is the predicted return for a stock using Model 1 when SMB = 3.30, HML = 1.25
and Rm-Rf = 5?
(A) 7.88%.
(B) 9.58%.
(C) 6.80%.

48 Which of the following statements least accurately describes one of the fundamental
multiple regression assumptions?
(A) The variance of the error terms is not constant (i.e., the errors are heteroskedastic.
(B) The independent variables are not random.
(C) The error term is normally distributed.

49 Henry Hilton, CFA, is undertaking an analysis of the bicycle industry. He hypothesizes

that bicycle sales (SALES) are a function of three factors: the population under 20 (POP),
the level of disposable income (INCOME), and the number of dollars spent on
advertising (ADV). All data are measured in millions of units. Hilton gathers data for the
last 20 years. Which of the follow regression equations correctly represents Hilton's
hypothesis?
(A) SALES =  + 1 POP + 2 INCOME + 3 ADV + .
(B) SALES =   1 POP  2 INCOME  3 ADV  .
(C) INCOME =  + 1 POP + 2 SALES + 3 ADV + .

50 One possible problem that could jeopardize the validity of the employment growth rate
model is multicollinearity. Which of the following would most likely suggest the
existence of multicollinearity?
(A) The F-statistic suggests that the overall regression is significant, however the
regression coefficients are not individually significant.
(B) The variance of the observations has increased over time.
(C) The Durbin—Watson statistic is significant.

51. Which of the following is least likely to be an assumption regarding linear regression?
(A) The variance of the residuals is constant.
(B) A linear relationship exists between the dependent and independent variables.
(C) The independent variable is correlated with the residuals.

Quantitative Methods 9 Multiple Regression

CFA
52. Based upon the information presented in the ANOVA table, what is the coefficient of
determination?
(A) 0.084, indicating that the variability of industry returns explains about 8.4% of
the variability of company returns.
(B) 0.839, indicating that company returns explain about 83.9% of the variability of
industry returns.
(C) 0.916, indicating that the variability of industry returns explains about 91.6% of
the variability of company returns.

53. Based upon her analysis, Carter has derived the following regression equation: Ŷ =
1.75 + 3.24X1.
The predicated value of the Y variable equals 50.50 if the:
(A) coefficient of the determination equals 15.
(B) predicated value of the dependent variable equals 15.
(C) predicated value of the independent variable equals 15.

54. Carter realize that although regression is a useful tool when analysing investments,
there are certain limitations. Carter made a list of points describing limitations that
Smith Brothers equality traders should be aware of when applying her research to their
investment decision.
 Point 1: Regression residuals may be homoskedastic.
 Point 2: Data from regressions relationship tends to exhibit parameter instability.
 Point 3: Regression residuals may exhibit autocorrelation.
 Point 4: The variance of the error term may change with one or more independent
variables.
When reviewing Carter's list, one of the Smith Brothers' equity traders points out that
not all of the points describe regression analysis limitations. Which of Carter's points
most accurately describes the limitations to regression analysis?
(A) Points 1, 2, and 3.
(B) Points 1, 3, and 4.
(C) Points 2, 3, and 4.

55. The percent of the variation in the fund’s that is explained by the regression is:
(A) 66.76%
(B) 61.78%
(C) 81.71%

Quantitative Methods 10 Multiple Regression

CFA
56. Suppose the Breusch-Godfrey statistic is 3.22. At a 5% level of significance, which of
the following is the most accurate conclusion regarding the presence of serial
correlation (at two lags) in the residuals?
(A) No, because the BG statistic is less than the critical test statistic of 3.55, we don't
have evidence of serial correlation.
(B) No, because the BG statistic is less than the critical test statistic of 3.49, we don't
have evidence of serial correlation.
(C) Yes, because the BG statistic exceeds the critical test statistic of 3.16, there is
evidence of serial correlation.

57. Gloucester subsequently revises the model to exclude the small cap index and finds that
the revised model has a RSS of 106.332. Which of the following statements is most
accurate? At a 5% level of significance, the test statistic.
(A) of 1.30 indicates that we cannot reject the hypothesis that the coefficient of small-
cap index is not significantly different from 0.
(B) of 4.35 indicates that we cannot reject the hypothesis that the coefficient of small-
cap index is significantly different from 0.
(C) of 13.39 indicates that we cannot reject the hypothesis that the coefficient of
small-cap index is significantly different from 0.

58. The best test for unconditional heteroskedasticity is:

(A) the Breusch-Godfrey test only.
(B) the Breusch-Pagan test only.
(C) neither the Durbin-Watson test nor the Breusch-Pagan test.

59. In the month of January, if both the small and large capitalization index have a zero
return, we would expect the fund to have a return equal to:
(A) 2.322.
(B) 2.799.
(C) 2.561.

60. Assuming (for this question only) that the F-test was significant but that the t-tests of
the independent variables were insignificant, this would likely suggest.
(A) serial correlation.
(B) multicollinearity.
(C) Conditional heteroskedasticity.

Quantitative Methods 11 Multiple Regression

CFA
61. Consider the following analysis of variance (ANOVA) table:
Source Sum of squares Degrees of Freedom Mean square
Regression 20 1 20
Error 80 40 2
Total 100 41
The F-statistic for the test of the fit of the model is closest to:
(A) 10.00.
(B) 0.10.
(C) 0.25.

62. Consider the following graph of residuals and the regression line from a time-series
regression:

These residuals exhibit the regression problem of:

(A) heteroskedasticity.
(B) autocorrelation.
(C) homoskedasticity.

63. Which of the following is least likely a method used to detect heteroskedasticity?
(A) Scatter plot.
(B) Breusch-Pagan test.
(C) Breusch-Godfrey test.

64. When pooling the samples over multiple economic environments in a multiple
regression model, which of the following errors is most likely to occur?
(A) Multicollinearity.
(B) Heteroskedasticity.
(C) Model misspecification.

Quantitative Methods 12 Multiple Regression

CFA
65. Concerning the assumptions of multiple regression, Grimbles is:
(A) incorrect to agree with Voiku's list of assumptions because one of the assumptions
is stated incorrectly.
(B) correct to agree with Voiku's list of assumptions.
(C) incorrect to agree with Voiku's list of assumptions because two of the assumptions
are stated incorrectly.

66. For which of the four hypotheses did Voiku incorrectly fail to reject the null, based on
the data given in the problem?
(A) Hypothesis 3.
(B) Hypothesis 2.
(C) Hypothesis 4.

67. The most appropriate decision with regard to the F-statistic for testing the null
hypothesis that all of the independent variables are simultaneously equal to zero at the
5 percent significance level is to:
(A) reject the null hypothesis because the F-statistic is larger than the critical F-value
of 2.66.
(B) fail to reject the null hypothesis because the F-statistic is smaller than the critical
F-value of 2.66.
(C) reject the null hypothesis because the F-statistic is larger than the critical F-value
of 3.19.

68. Regarding Voiku’s calculations of R2 and the standard error of estimate, she is:
(A) incorrect in her calculation of the unadjusted R2 but correct in her calculation of
the standard error of estimate.
(B) incorrect in her calculation of both the unadjusted R2 and the standard error of
estimate.
(C) correct in her calculation of the unadjusted R2 but incorrect in her calculation of
the standard error of estimate.

69. The multiple regressions, as specified, most likely suffers form:

(A) heteroskedasticity.
(B) serial correlation of the error terms.
(C) multicollinearity.

70. A 90 percent confidence interval for the coefficient on GDP is:

(A) 0.5 to 22.9.
(B) –1.5 to 20.2.
(C) –1.9 to 19.6.

Quantitative Methods 13 Multiple Regression

CFA
71. An analyst is trying to determine whether fund return performance is persistent. The
analyst divides funds into three groups based on whether their return performance was
in the top third (group 1), middle third (group 2), or bottom third (group 3) during the
previous year. The manager then creates the following equation: R = a + b1D1 + b2D2 +
b3D3 + , where R is return premium on the fund (the return minus the on the S & P
500 benchmark) and Di is equal to 1 if the fund is group i.
Assuming no other information, this equation will suffer from:
(A) multicollinearity
(B) serial correlation.
(C) heteroskedasticity.

72. Henry Hilton, CFA, is understanding an analysis of the bicycle industry. He hypothesizes
that bicycle sales (SALES) are a function of three factors; the population under 20 (POP),
the level of disposable income (INCOME), and the number of dollars spent on
advertising (ADV), All data are measured in millions of units. Hilton gathers data for the
last 20 years and estimates the following equation
(Standard errors in parentheses):
SALES = + 0.004 POP + 1.031 INCOME + 2.002 ADV
(0.005) (0.337) (2.312)
The critical t-statistic for a 95% confidence level is 2.120. Which of the independent
variables is statistically different from zero at the 95% confidence level?
(A) ADV only.
(B) INCOME only.
(C) INCOME and ADV.

73. An analyst runs a regression of monthly values-stock returns on five independent

variables over 48 months. The total sum of squares is 430, and the sum of squared
errors is 170. Test the null hypothesis at the 2.5% significance level that all five of the
independent variables are equal to zero.
(A) Not rejected at 2.5% or 5.0% significance.
(B) Rejected at 2.5% significance and 5% significance.
(C) Rejected at 5% significance only.

74. An analyst is trying to estimate the beta for a fund. The analyst estimates a regression
equation in which the fund returns are the dependent variable and the Wilshire 5000 is
the independent variable, using monthly data over the past five years. The analyst finds
that the correlation between the square of the residuals of the regression and the
Wilshire 5000 is 0.2. Which of the following is most accurate, assuming a 0.05 level of
significance? There is:
(A) evidence of level serial correlation but not conditional heteroskedasticity in the
regression equation.
(B) evidence of conditional heteroskedasticity but not serial correlation in the
regression equation.
(C) no evidence that there is conditional heteroskedasticity or serial correlation in the
regression equation.
Quantitative Methods 14 Multiple Regression
CFA
75. Which of the following statements regarding heteroskedasticity is least accurate?
(A) Conditional heteroskedasticity can be detected using the Breusch-Pagan chi-
square statistic.
(B) When not related to independent variables, heteroskedasticity does not pose any
major problems with the regression.
(C) Heteroskedasticity only occurs in cross-sectional regressions.

76. which of the following statements most accurately intercepts the following regression
results at the given significance level?
Variable p-value
Intercept 0.0201
X1 0.0284
X2 0.0310
X3 0.0143
(A) The variables X1 and X2 are statistically significantly different from zero at the 2%
significance level.
(B) The variable X3 is statistically significantly different from zero at the 2%
significance level.
(C) The variable X2 is statistically significantly different from zero at the 3%
significance level.

77. In a one-side test and a 1% level significance, which of the following coefficients is
significantly difference from zero?
(A) The intercept and the coefficient on In (market value) only.
(B) The intercept and the coefficient on In (no. of analysts) only.
(C) The coefficient on In(no. of Analysts) only.

78. The 95% confidence interval (use a t-stat of 1.96 for this equation only) of the
estimated coefficient for the independent variable Ln (Market Value) is closest to:
(A) 0.011 to 0.001.
(B) –0.018 to –0.036.
(C) 0.014 to –0.009.

79. If the number of analysts on NGR Corp. were to double to 4, the change in the forecast
of NGR would be closest to?
(A) –0.035.
(B) –0.055.
(C) –0.019.

Quantitative Methods 15 Multiple Regression

CFA
80. Base on a R2 calculated from the information in Table 2, the analyst should conclude
that the number of analysts and In (market value) of the firm explain:
(A) 15.6 of the variation in returns.
(B) 18.4 of the variation in returns.
(C) 84.4% of the variation in returns.

81. What is the F-statistic from the regression? And, what can be concluded from its value
at a 1% level of significance?
(A) F = 1.97, fail to reject a hypothesis that both of the slope coefficient are equal to zero.
(B) F = 5.80, reject a hypothesis that both of the slope coefficients are equal to zero.
(C) F = 17.00, reject a hypothesis that both of the slope coefficients are equals to zero.

82. Upon further analysis, Turner concludes that multicollinearity is a problem. What might
have prompted this further analysis and what is intuition the conclusion?
(A) At least one of the t-statistics was not significant, the F-statistic was not
significant, and a positive relationship between the number of analysts and the
size of the firm would be expected.
(B) At least one of the t-statistics was not significant, the F-statistic was not
significant, and a positive relationship between the number of analysts and the
size of the firm would be expected.
(C) At least one of the t-statistics was not significant, the F-statistic was significant,
and an intercept not significantly different from zero would be expected.

83. When interpreting the results of a multiple regression analysis, which of the following
terms represents the value of the dependent variable when the independent variables
are all equal to zero?
(A) Intercept term.
(B) Slope coefficient.
(C) p-value.

84. Consider the following estimated regression equation, with the standard errors of the
slope coefficient as noted:
Salesi = 10.0 + 1.25 R&Di + 1.01 ADVi – 2.0 COMPi + 8.0 CAPi
Where the standard error for the estimated coefficient on R&D is 0.45, the standard
error for the estimated coefficient on ADV is 2.2, the standard error for the estimated
coefficient on COMP is 0.63, and the standard error for the estimated coefficient on
CAP is 2.5.
The equation was estimated over 40 companies. Using a 5% level of significance, which
of the estimate coefficient are significantly different from zero?
(A) ADV and CAP only.
(B) R&D, COMP, and CAP only.
(C) R&D, ADV, COMP, and CAP.

Quantitative Methods 16 Multiple Regression

CFA
85. Alex Wade, CFA, is analyzing the result of a regression analysis comparing the
performance of gold stocks versus a board equity market index. Wade believes that first
lag serial correlation may be present and, in order to prove his theory, should use which
of the following methods to detect its presence?
(A) The Hansen method.
(B) The Breusch-Pagan test.
(C) The Durbin-Watson statistic.

86. Consider the following model of earnings (EPS) regressed against dummy variables for
the quarters:
EPSt =  + 1Q1t + 2Q2t + 3Q3t
Where:
EPSt is a quarterly observation of earnings per shares
Q1t takes on a value of 1 if period t is the second quarter, 0 otherwise
Q2t takes on a value of 1 period t is the third quarter, 0 otherwise
Q3t take on a value of 1 if period t is the fourth quarter, 0 otherwise
Which of the following statements regarding this model is most accurate? The:
(A) coefficient on each dummy tells us about the difference in earnings per share
between the respective quarter and the one left out (first quarter in this case).
(B) EPS for the first quarter is represented by the residual.
(C) Significance of the coefficient cannot be intercepted in the case of dummy
variables.

87. Which of the following questions is least likely answered by using a qualitative
dependent variables?
(A) Based on the following executive-specific and company-specific variables, how
many shares will be acquired through the exercise of executive stock options?
(B) Based on the following subsidiary and competition variables, will company XYZ
divest itself of a subsidiary?
(C) Based on the following company-specific financial ratios, will company ABC enter
bankruptcy?

88. A high-yield bond analyst is trying to develop an equation using financial ratios to
estimate the probability of a company defaulting on its bonds. A technique that can be
used to develop this equation is:
(A) Dummy variable regression.
(B) Logistic regression model.
(C) Multiple linear regression adjusting for heteroskedasticity.

Quantitative Methods 17 Multiple Regression

CFA
89. Consider the following estimated regression equation, with calculated t-statistics of the
estimates as indicated:
AUTOt = 10.0 + 1.25 Plt + 1.01 TEENt – 2.0 INSt
With a PI calculated t-statistic of 0.45, a TEEN calculated t-statistic of 2.2, and an INS
calculated t-statistic of 0.63.
The equation was estimated over 40 companies. The predicated value of AUTO if PI is 4,
TEEN is 0.30, and INS = 0.6 is closest to:
(A) 14.10
(B) 17.50.
(C) 14.90.

90. Which of the following statements regarding heteroskedasticity is least accurate?

(A) The assumption of linear regression is that the residuals are heteroskedastic.
(B) Heteroskedasticity may occur in cross-sectional or time-series analyses.
(C) Heteroskedasticity results in an estimated variance that is too small, and therefore
affects statistical inference.

91. When two or more of the independent variables in a multiple regression are correlated
with each other, the condition is called:
(A) conditional heteroskedasticity.
(B) multicollinearity.
(C) serial correlation.

92. Consider the following regression equation:

Salesi = 10.0 + 1.25 R&Di + 1.0 ADVi – 2.0 COMPi + 8.0 CAPi
Where sales is dollar sales in millions, R&D research and development expenditures in
millions, ADV is dollar amount spent on advertising in millions, COMP is the number of
competitors in the industry, and CAP is the capital expenditures for the period in
millions of dollars.
Which of the following is NOT a correct interception of this regression information?
(A) If a company spends $1 million more on capital expenditures (holding everything
else constant), Sales are expected to increase by $8.0 million.
(B) If R & D and advertising expenditures are $1 million each, there are 5 competitors,
and capital expenditures are $2 million, expected Sales are $8.25 million.
(C) One more competitors will mean $2 million less in Sales (holding everything else
constant).

Quantitative Methods 18 Multiple Regression

CFA
93. Henry Hilton, CFA, is understanding an analysis of the bicycle industry. He hypothesizes
that bicycle sales (SALES) are a function of three factors: the population under 20 (POP),
the level of disposable income (INCOME), and the number of dollars spent on
advertising (ADV), All data are measured in millions of units. Hilton gathers data for the
last 20 years and estimates the following equation
(Standard errors in parentheses):
SALES = 0.000 + 0.004 POP + 1.031 INCOME + 2.002 ADV
(0.113) (0.005) (0.337) (2.312)
For next year, Hilton estimates the following parameters: (1) the population under 20
will be 120 million, (2) disposable income will be $300,000,000, and (3) advertising
expenditures will be $100,000,000. Based on these estimated and the regression
equation, what are predicated sales for the industry for next year?
(A) $509,980,000.
(B) $557,143,000.
(C) $656,991,000.

94. If GDP rises 2.2% and the price of fuels falls $0.15, Baltz’s model will predict Company
sales to be (in $ millions) closest is:
(A) $82.00
(B) $128.00.
(C) $206.00.

95. Baltz proceeds to test the hypothesis that none of the independent variables has
significant explanatory power. He concludes that, at 5% level of significance.
(A) all of the independent variables have explanatory power, because the calculated F-
statistic exceeds its critical value.
(B) none of the independent variable has explanatory power, because the calculated
F-statistic does not exceed its critical value.
(C) at least one of the independent variables has explanatory power, because the
calculated F-statistic exceeds its critical value.

96. Presence of conditional heteroskedasticity is least likely to affect the:

(A) Computed F-statistic
(B) coefficient estimates.
(C) computed t-statistic.

Quantitative Methods 19 Multiple Regression

CFA
97. An analyst is estimating whether company sales is related to three economic variables.
The regression exhibits conditional heteroskedasticity, serial correlation, and
multicollinearity. The analyst uses White and Newey-West standard errors. Which of the
following is most accurate?
(A) The regression will still exhibit serial correlation and multicollinearity, but the
heteroskedasticity problem will be solved.
(B) The regression will still exhibit heteroskedasticity and multicollinearity, but the
serial correlation problems will be solved.
(C) The regression will still exhibit multicollinearity, but the heteroskedasticity and
serial correlation problems will be solved.

98. A regression with three independent variables have VIF values of 3, 4 and 2 for the
first, second, and third independent variables, respectively. Which of the following
conclusions is most appropriate?
(A) Multicollinearity does not seem to be problem with the model.
(B) Only variable two has a problem with multicollinearity.
(C) Total VIF of 9 indicates a serious multicollinearity problem.

99. The management of a large restaurant chain believes that revenue growth is dependent
upon the month of the year. Using a standard 12 month calendar, how many dummy
variables must be used in a regression model that will test whether revenue growth
differs by month?
(A) 11.
(B) 13.
(C) 12.

100. Which of the following statements regarding the R2 is least accurate?

(A) The F-statistic for the test of the fit of the model is the ratio of the mean squared
regression to the mean squared error.
(B) The R2 is the ratio of the unexplained variation to the explained variation of the
dependent variable.
(C) The R2 of a regression will be greater than or equal to the adjusted-R2 for the
same regression.

101. Which of the following is least likely to result in misspecification of a regression model?
(A) Transforming a variable.
(B) Inappropriate variable form.
(C) Omission of an important independent variable.

Quantitative Methods 20 Multiple Regression

CFA
102. In regard to their conversion about the regression equation:
(A) Brent’s statement is correct; Johnson’s statement is incorrect.
(B) Brent’s statement is incorrect; Johnson’s statement is correct.
(C) Brent’s statement is correct; Johnson’s statement is correct.

103. Using data from the past 20 quarters, Brent calculates the t-statistic for marketing
expenditures to be 3.68 and the t-statistic for salespeople at 2.19. At a 5% significance
level, the two-tailed critical values are tc = +/– 2.127. This most likely indicated that:
(A) the t-statistic has 18 degrees of freedom.
(B) the null hypothesis should not be rejected.
(C) both independent variables are statistically significant.

104. Brent calculated that the sum of squared errors (SSE) for the variables is 267. The
means squared error (MSE) would be:
(A) 14.055.
(B) 15.706.
(C) 17.831.

105. Brent is trying to explain the concept of the standard error of estimate (SEE) to Johnson.
In this explanation, Brent makes three points about the SEE.
• Point 1: The SEE is the standard deviation of the differences between the
estimated values for the independent variable and the actual observations for the
independent variable.
• Point 2: Any violation of the basic assumptions of a multiple regressions model is
going to affect the SEE.
• Point 3. If there is a strong relationship between the variables and the SEE is
small, the individual estimation errors will also be small.
How many of Brent’s points are most accurate?
(A) 1 of Brent’s points are correct.
(B) All 3 of Brent’s points are correct.
(C) 2 of Brent’s points are correct.

106. Assuming that next year’s marketing expenditures are $3,500,000 and there are five
salespeople, predicated sales for Mega Flowers should will be:
(A) $24,000,000.
(B) $11,600,000.
(C) $24,200,000.

Quantitative Methods 21 Multiple Regression

CFA
107. Brent would like to further investigate whether at least one of the independent variables
can explain a significant portion of the variation of the dependent variable. Which of the
following methods would be best for Brent to use?
(A) The multiple coefficient of determination.
(B) The F-statistic.
(C) An ANOVA table.

108. May jones estimated a regression that produced the following analysis of variance
(ANOVA) table:
Source Sum of square Degrees of freedom Mean square
Regression 20 1 20
Error 80 40 2
Total 100 41
The values of R2 and the F-statistic for joint test of significance of all the slope
coefficients are:
(A) R2 = 0.20 and F = 10.
(B) R2 = 0.25 and F = 0.909.
(C) R2 = 0.25 and F = 10.

109. According to the model and the data for the Chicago metropolitan area, the forecast of
generator sales is:
(A) $55 million above average.
(B) $35.2 million above average.
(C) $65 million above average.

110. Williams proceeds to test the hypothesis that none of the independent variables has
significant explanatory power. Using the joint F-test for the significance of all slope
coefficients, at a 5% level of significance:
(A) all of the independent variables have explanatory power.
(B) none of the independent variables has explanatory power.
(C) at least one of the independent variables has explanatory power.

111. With respect to testing the validity of the model’s results. Williams may wish to perform:
(A) a Breusch-Godfrey test, but not a Breusch-Pagan test.
(B) both a Breusch-Godfrey test and a Breusch-Pagan test.
(C) a Breusch-pagan test, but not Breusch-Godfrey.

Quantitative Methods 22 Multiple Regression

CFA
112. When Williams ran the model, the computer said the R2 is 0.233. She examines the
other output and conclusion that this is the:
(A) neither the unadjusted nor adjusted R2 value, nor the coefficient of correlation.
(B) unadjusted R2 value.
(C) adjusted R2 value.

113. In preparing and using this model, Williams has least likely relied on which of the
following assumptions?
(A) There is a linear relationship between the independent variables.
(B) The disturbance or error term is normally distributed.
(C) The residuals are homoscedastic.

114. Jason Fye, CFA, wants to check for seasonality in monthly stock returns (i.e., the January
effect) after controlling for market cap and systematic risk. The type of model that Fye
would most appropriately select is:
(A) Multiple regression model.
(B) logistic regression model.
(C) Neither multiple regression nor logistic regression.

115. Using the regression model represented in Exhibit 1, What is the predicated number of
housing starts for 20X7.
(A) 1,394,420
(B) 1,751,000
(C) 1,394

116. Which of the following statements best describes the explanatory power of the
estimated regression?
(A) The residual standard error of only 0.3 indicated that the regression equation is a
good fit for the sample data.
(B) The large F-statistic indicates that both independent variables help explain
changes in housing starts.
(C) The independent variables explain 61.58% of the variation in housing starts.

117. Which of the following is the least appropriate statement in relation to R-square and
adjusted R-square:
(A) Adjusted R-square is a value between 0 and 1 can be interpreted as a percentage.
(B) R-square typically increase when new independent variables are added to the
regression regardless of their explanatory power.
(C) Adjusted R-square decrease when the added independent variable adds little
value to the regression model.

Quantitative Methods 23 Multiple Regression

CFA
118. Which of the following statements regarding the results of a regression analysis is least
accurate?
The:
(A) slope coefficient in a multiple regression is the value of the dependent variable for
a given value of the independent variable.
(B) slope coefficient in a multiple regression is the change in the dependent variable
for a one-unit change in the independent variable, holding all other variables
constants.
(C) slope coefficient in the multiple regression are referred to as partial betas.

119. Using the regression model developed, the closest prediction of sales for December
20X6 is:
(A) $36,000
(B) $44,000.
(C) $35,000.

120. Will Stumper conclude that the housing starts coefficient is statistically different from
zero and how will he interpret it at the 5% significance level:
(A) different from zero; sales will rise by $100 for every 23 house starts.
(B) different from zero; sales will rise by $23 for every 100 house starts.
(C) not different from zero; sales will rise by $0 for every 100 house starts.

121. Is the regression coefficient to changes in mortgage interest rates different from zero at
the 5 percent level of significance?
(A) yes, because 2.6 > 2.23.
(B no, because 2.6 < 2.62.
(C) yes, because 2.6 > 1.98.

122. In this multiple regression, the F-statistic indicates the:

(A) the joint significance of the independent variables.
(B) deviation of the estimated value from the actual values of the dependent variable.
(C) degree of correlation between the independent variables.

123. The regression statistics above indicate that for the period under study, the
independent variables (housing starts, mortgage interest rate) together explained
approximately what percentage of the variation is the dependent variable (sales)?
(A) 9.80.
(B) 67.00.
(C) 77.00.

Quantitative Methods 24 Multiple Regression

CFA
124. In this multiple regression, if Stumper discovers that the residuals exhibit positive serial
correlation, the most likely effect is:
(A) standard errors are too low but coefficient estimate is consistent.
(B) standard errors are too high but coefficient estimate is consistent.
(C) standard errors are not affected but coefficient estimate is inconsistent.

125. Which of the following tests is least likely to be used to detect autocorrelation?
(A) Durbin-Watson.
(B) Breusch-Godfrey.
(C) Breusch-Pagan.

126. One of the most popular ways to correct heteroskedasticity is to:

(A) Improve the specification of the model.
(B) adjust the standard errors.
(C) use robust standard errors.

127. If a regression equation shown that no individual t-tests are significance, but the
F-statistic is significance, the regression probably exhibits:
(A) serial correlation.
(B) multicollinearity.
(C) heteroskedasticity.

128. Consider the following estimated regression equation, with standard errors of the
coefficients as indicated:
Salesi = 10.0 + 1.25 R&Di + 1.0 ADVi – 2.0 COMPi + 8.0 CAPi
Where the standard error for R&D is 0.45, the standard error for ADV is 2.2, the
standard error for COMP 0.63, and the standard error for CAP is 2.5.
Sales are in millions of dollars. An analyst is given the following predictions on the
independent variables: R&D = 5, ADV = 4, COMP = 10, and CAP = 40.
The predicated level of sales is closest to:
(A) $310.25 million.
(B) $300.25 million.
(C) $320.25 million.

129. Jacob Warner, CFA, is evaluating a regression analysis recently published in a trade
journal that hypothesizes that the annual performance of the S&P 500 stock index can
be explained by movements in the Federal Funds rate and the U.S. Producer Price index
(PPI). Which of the following statements regarding his analysis is most accurate?

Quantitative Methods 25 Multiple Regression

CFA
(A) If the t-value of a variable is less than the significance level, the null hypothesis
should be rejected.
(B) If the p-value of a variable is less than the significance level, the null hypothesis
cannot be rejected.
(C) If the p-value of a variable is less than the significance level, the null hypothesis
can be rejected.

130. What is most likely represented by the Y intercept of the regression?

(A) The intercept is not a driver of returns, only the independent variables.
(B) The drift of a random walk.
(C) The return on a particular trading day.

131. What can be said of the overall explanatory power of the model at the 5% significance?
(A) The coefficient of determination for the above regression is significantly higher
than the standard error of the estimate, and therefore there is value to calendar
trading.
(B) There is no value to calendar trading.
(C) There is value to calendar trading.

132. The test mentioned by Jessica is known as the:

(A) Breusch-Pagan, which is a two-tailed test.
(B) Durbin-Watson, which is a two-tailed test.
(C) Breusch-Pagan, which is a one-tailed test.

133. Are Jessica and her son Jonathan correct in terms of the method used to correct for
heteroskedasticity and the likely effects?
(A) Both are correct.
(B) One is correct.
(C) Neither is correct.

134. During the course of a multiple regression analysis, an analyst has observed several
items that she believes may render incorrect conclusions. For example, the coefficient
standard errors are too small, although the estimated coefficient are accurate. She
believes that these small standard error terms will result in the computed t-statistics
being too big, resulting in too many Type I errors. The analyst has most likely observed
which of the following assumption violations in her regression analysis?
(A) Positive serial correlations.
(B) Multicollinearity.
(C) Homoskedasticity.

Quantitative Methods 26 Multiple Regression

CFA
135. Consider the following regression equation:
Salesi = 20.5 + 1.5 R&Di + 2.5 ADVi – 3.0 COMPi
where Sales is dollar sales in millions, R&D is research and development expenditures in
millions, ADV is dollar amount spent on advertising in millions, and COMP is the
number of competitors in the industry.
Which of the following is NOT a correct interpretation of this regression information?
(A) If a company spends $1 more on R&D (holding everything else constant), sales are
expected to increase by $ 1.5 million.
(B) One more competitor will mean $3 million less in sales (holding everything else
constant).
(C) If R&D and advertising expenditures are $1 million each and there are
5 competitors, expected sales are $9.5 million.

136. Using the regression model developed, the closest predication sales for December
20X6 is:
(A) $44,000.
(B) $36,000.
(C) $55,000.

137. Will Jack conclude that the housing starts coefficient is statistically different from zero
and how will he intercept it at the 5% significance level?
(A) Different from zero; sales will rise by $100 for every 23 house starts.
(B) Different from zero’ sales will rise by $23 for every 100 house starts.
(C) Not different from zero; sales will rise by $0 for every 100 house starts.

138. In this multiple regression, the F-statistic indicates the:

(A) the joint significance of the independent variables.
(B) deviation of the estimated values from the actual values of the dependent variable.
(C) degree of correlation between the independent variables.

139. The regression statistics indicate that for the period under study, the independent
variables (housing starts, mortgage interest rate) together explain approximately what
percentage of the variation in the dependent variable (sales)?
(A) 77.00.
(B) 9.80.
(C) 67.00.

Quantitative Methods 27 Multiple Regression

CFA
140. For this question only, assume that the regression of squared residuals on the
independent variables has R2 = 11%. At a 5% level significance, which of the following
conclusions is most accurate?
(A) Because the critical value is 3.84, we reject the null hypothesis of no conditional
heteroskedasticity.
(B) With a test statistics of 13.53, we can conclude the presence of conditional
heteroskedasticity.
(C) With a test statistics of 0.22, we cannot reject the null hypothesis of no
conditional heteroskedasticity.

141. Wilson estimated a regression that produced the following analysis of variance (ANOVA)
table:
Source Sum of squares Degrees of freedom Mean square
Regression 100 1 100.0
Error 300 40 7.5
Total 400 41
2
The values of R and the F-statistic to test the null hypothesis that slope coefficient on
all variables are equal to zero are:
(A) R2 = 0.20 and F = 13.333.
(B) R2 = 0.25 and F = 13.333.
(C) R2 = 0.25 and F = 0.930.

142. Jill Wentraub is an analyst with the retail industry. She is modelling a company’s sales
over time and has noticed a quarterly seasonal pattern. If she includes dummy variables
to present the seasonally component of the sales she must use:
(A) one dummy variables.
(B) four dummy variables.
(C) three dummy variables.

143. How many of the three independent variables (not including the intercept term) are
statistically significance in explaining quarterly stock returns at the 5.0% level?
(A) All there are statistically significant.
(B) Two of the three are statistically significant.
(C) One of the three is statistically significant.

144. Can the null hypothesis that the GDP growth coefficient is equal to 3.50 be rejected at
the 1.0% confidence level versus the alternative that it is not equal to 3.50? The null
hypothesis is:
(A) not rejected because the t-statistic is equal to 0.92.
(B) rejected because the t-statistic is less than 2.617.
(C) accepted because the t-statistic is less than 2.617.

Quantitative Methods 28 Multiple Regression

CFA
145. The percentage of the total variation in quarterly stock returns explained by the
independent variables closest to:
(A) 32%.
(B) 47%.
(C) 42%.

146. According to the Durbin-Watson statistic, there is:

(A) significant positive serial correlation in the residuals.
(B) significant heteroskedasticity in the residuals.
(C) no significant positive serial correlation in the residuals.

147. What is predicated quarterly stock return, given the following forecasts?
 Employment growth = 2.0%
 GDP growth = 1.0%
 Private investment growth = –1.0%
(A) 4.4%.
(B) 5.0%.
(C) 4.7%.

148. What is the standard error of the estimate?

(A) 0.81.
(B) 1.71.
(C) 1.31.

149. Using the regression model represented in Exhibit 1, what is the predicated number of
housing starts for 20X7?
(A) 1,394,420.
(B) 1,751,000.
(C) 1,394.

150. Which of the following statements best describes the explanatory power of the
estimated regression?
(A) The independent variables explain 61,58% of the variation in housing starts.
(B) The large F-statistic indicates that both independent variable help explain changes
in housing starts.
(C) The residual standard error of only 0.3 indicates that the regression is a good fit
for the sample data.

Quantitative Methods 29 Multiple Regression

CFA
151. Which of the following is the least appropriate statement in relation to R-square and
adjusted R-square?
(A) Adjusted R-square decreases when the added independent variable adds little
value to the regression model.
(B) R-square typically increases when new independent variables adds to the
regression.
(C) Adjusted R-square can be higher than the coefficient of determination for a model
with a good fit.

152. What is the correct interpretation of the coefficient of closed in the first regression?
(A) If a model is closed to new investors, the expected excess fund return is 1.65%.
(B) A closed fund is estimated to have an extra returns of 1.65% relative to funds that
are not closed.
(C) A closed fund is likely to generate a return of 1.65%.

153. To check for only the outliers in the sample, Lee should most appropriately use:
(A) leverage.
(B) Cook’s D.
(C) Studentized residuals.

154. Which observations, when excluded, cause a significance change to model coefficients?
(A) Observation 10 and 19.
(B) Observation 1, 10, and 11.
(C) Observation19.

155. What is the change probability of fund closure for a 1% increase in Ln(assets under
management)?
(A) 5.08%
(B) 2.33%.
(C) 4.83%.

156. Which of the following statement regarding serial correlation that might be encountered
in regression analysis is least accurate?
(A) Serial correlation does not affect consistency of regression coefficients.
(B) Positive serial correlation and heteroskedasticity can both lead to Type I errors.
(C) Serial correlation occurs least often with the time series data.

Quantitative Methods 30 Multiple Regression

CFA
157. Which of the following is least likely a method of detecting serial correlation?
(A) The Breusch-Godfrey test.
(B) A scatter plot of the residuals over time.
(C) The Breusch-Pagan test.

158. A multiple regression model has included independent variables that are not linearly
related to the dependent variable. The model is most likely misspecificed due to:
(A) incorrect data pooling.
(B) incorrect variable form.
(C) incorrect variable scaling.

Quantitative Methods 31 Multiple Regression