Term
Best test for unconditional heteroskedasticity is:
Best test for conditional heteroskedasticity is:
Which is more serious than the other? |
|
Definition
No tests for unconditional.
Breusch-Pagan for conditional
Conditional considered more serious than unconditional
The Durbin-Watson test is for serial correlation. The Breusch-Pagan test is for conditional heteroskedasticity; it tests to see if the size of the independent variables influences the size of the residuals. Although tests for unconditional heteroskedasticity exist, they are not part of the CFA curriculum, and unconditional heteroskedasticity is generally considered less serious than conditional heteroskedasticity. |
|
|
Term
When the F-test and the t-tests conflict: |
|
Definition
Multicollinearity is indicated |
|
|
Term
Three-factor arbitrage pricing theory (APT) model expected return (formula) |
|
Definition
=RFR + S1RFP1 + S2RFP2 + ...
Given a three-factor arbitrage pricing theory APT model, what is the expected return on the Freedom Fund?
The factor risk premiums to factors 1, 2, and 3 are 10%, 7% and 6%, respectively.
The Freedom Fund has sensitivities to the factors 1, 2, and 3 of 1.0, 2.0 and 0.0, respectively.
The risk-free rate is 6.0%.
|
A)
|
33.0%.
|
|
B)
|
24.0%.
|
|
C)
|
30.0%.
|
Your answer: A was incorrect. The correct answer was C) 30.0%.
The expected return on the Freedom Fund is 6% + (10.0%)(1.0) + (7.0%)(2.0) + (6.0%)(0.0) = 30.0%. |
|
|
Term
Appopriate # of dummy variables to be used in a regression model |
|
Definition
|
|
Term
|
Definition
Spurious correlation occurs when the analysis erroneously indicates a relationship between two variables when none exists. |
|
|
Term
F-statistic (formula) and formulas of Num/Dom |
|
Definition
F = MSR / MSE
The F-statistic is equal to the ratio of the mean squared regression to the mean squared error.
MSR = RSS / k
MSE = SSE / (n-k-1) |
|
|
Term
Random walk time series (formula)
For a random walk, the long-run mean is (answer and formula) |
|
Definition
xt = xt-1 + et
Undefined. The slope coefficient is one, b1=1, and that is what makes the long-run mean undefined:
mean = b0/(1-b1) |
|
|
Term
Mean-reverting level for AR (1) |
|
Definition
|
|
Term
Relationship between two variables when there is a standard error of estimate (SEE) that is high relative to total variability |
|
Definition
Weak. The SEE is the standard deviation of the error terms in the regression, and is an indicator of the strength of the relationship between the dependent and independent variables. The SEE will be low if the relationship is strong and conversely will be high if the relationship is weak. |
|
|
Term
Assumptions of linear regression:
(1) Expected value/mean of the residuals is ______
(2) residuals are _____ distributed
(3) ______ variance |
|
Definition
(1) zero
(2) independently
(3) constant |
|
|
Term
An indication of multicollinearity is when: |
|
Definition
(1) if two ind. variables only--high corr. between the two
(2) high R2 and significant F-stat but not significant coefficients on ind. vars. |
|
|
Term
The R2 is the ratio of the: |
|
Definition
explained variation to the total variation |
|
|
Term
The Durbin-Watson test is used to: |
|
Definition
detect serial correlation |
|
|
Term
The Breusch-Pagan test is used to: |
|
Definition
detect heteroskedasticity |
|
|
Term
Estimation of first differences:
Salest = b0 + b1 Sales t-1+ εt
What is the specification of the model if first differences are used? |
|
Definition
(Salest - Salest-1)= b0 + b1 (Sales t-1 - Sales t-2) + εt. |
|
|
Term
A time series x that is a random walk with a drift is best described as: |
|
Definition
|
|
Term
The standard error of the estimate (SEE) in a regression is the (def) (2 formulas) (smaller/larger is better fit?) |
|
Definition
residuals of the regression. It is the standard deviation of the residuals.
SEE = √[SSE/(n-2)]
SEE = sqrt(MSE)
Smaller is better fit. |
|
|
Term
When is there economic significance to a strategy?
Market inefficiency is avaliable when: |
|
Definition
When there are abnormal returns AND they can cover transaction costs
When excess returns are available after covering transaction costs
|
|
|
Term
A simple linear regression equation had a coefficient of determination (R2) of 0.8. What is the correlation coefficient between the dependent and independent variables and what is the covariance between the two variables if the variance of the independent variable is 4 and the variance of the dependent variable is 9? |
|
Definition
The correlation coefficient is the square root of the R2, r = 0.89.
To calculate the covariance multiply the correlation coefficient by the product of the standard deviations of the two variables:
COV = 0.89 × √4 × √9 = 5.34 |
|
|
Term
The root mean squared error (RMSE) criterion is used to (formula + interpret values): |
|
Definition
compare the accuracy of autoregressive models in forecasting out-of-sample values.
RMSE = sqrt(average squared error)
The model with the smallest RMSE is the preferred model.
The RMSE for Model 1 is √10.429 = 3.23, while the RMSE for Model 2 is √11.642 = 3.41. Since Model 1 has the lowest RMSE, that is the one Zox should conclude is the most accurate. |
|
|
Term
The coefficient of determination / R2 (definition + formula) |
|
Definition
the percentage of total variation in the dependent variable explained by the independent variable
R2 = (RSS / SST) |
|
|
Term
|
Definition
a qualitative dependant variable which is based on a normal distribution |
|
|
Term
|
Definition
a qualitative dependant variable which is based on the logistic distribution |
|
|
Term
|
Definition
returns a qualitative dependant variable based on a linear relationship that can be used for ranking or classification into discrete states |
|
|
Term
When utilizing a proxy for one or more independent variables in a multiple regression model, what error is likely to occur? |
|
Definition
Model misspecification
By using a proxy for an independent variable in a multiple regression analysis, there is some degree of error in the measurement of the variable. |
|
|
Term
The mean regression sum of squares (MSR) formula: |
|
Definition
RSS/k
The regression sum of squares divided by the number of independent variables
|
|
|
Term
Residual sum of squares (RSS) |
|
Definition
RSS = SST - SSE
The residual sum of squares is the difference between the total sum of squares and the regression sum of squares
|
|
|
Term
Reject null hypothesis of no significance if: |
|
Definition
|t| > critical t
-or-
p-value < alpha |
|
|
Term
The standard error of estimate (SEE) |
|
Definition
|
|
Term
Positive serial correlation |
|
Definition
Positive serial correlation is the condition where a positive regression error in one time period increases the likelihood of having a positive regression error in the next time period. The residual terms are correlated with one another, leading to coefficient error terms that are too small. |
|
|
Term
Durbin-Watson test (formula + purpose) |
|
Definition
The Durbin-Watson statistic tests for serial correlation. For large samples, the Durbin-Watson statistic is approximately equal to:
2 * (1 - r)
r = sample correlation between the regressions residuals from one period and the residuals from the previous period
If r < (lower) DW value on chart, REJECT H0 of no corr.
For 90 observations, use 90 df for DW
|
|
|
Term
t-test for correlation(r) (formula) and df calculation |
|
Definition
t =
[r*(sqrt(n-2)) / (sqrt(1-r2))]
df = n - 2 |
|
|
Term
David Wellington, CFA, has estimated the following log-linear trend model: LN(xt) = b0 + b1t + εt. Using six years of quarterly observations, 2001:I to 2006:IV, Wellington gets the following estimated equation: LN(xt) = 1.4 + 0.02t. The first out-of-sample forecast of xt for 2007:I is closest to: |
|
Definition
The first out-of-sample forecast of xt for 2007:I is closest to:
|
A)
|
1.88.
|
|
B)
|
6.69.
|
|
C)
|
4.14.
|
Your answer: A was incorrect. The correct answer was B) 6.69.
Wellington’s out-of-sample forecast of LN(xt) is 1.9 = 1.4 + 0.02 × 25, and e1.9 = 6.69. |
|
|
Term
Dickey-Fuller Test tests for (+ formula): |
|
Definition
nonstationarity; which uses a modified t-statistic
The Dickey-Fuller test estimates the equation:
(xt – xt-1) = b0 + (b1 - 1) * xt-1 + et and tests if H0: (b1 – 1) = 0.
Using a modified t-test, if it is found that (b1–1) is not significantly different from zero, then it is concluded that b1 must be equal to 1.0 and the series has a unit root.
|
|
|
Term
The Hansen Method (use + when to use) |
|
Definition
The Hansen method adjusts for problems associated with both serial correlation and heteroskedasticity
Used if DW tests finds positive autocorrelation |
|
|
Term
If there is only one independent variable, ____ cannot be a problem |
|
Definition
|
|
Term
Serial correlation always impacts the statistical inference about: |
|
Definition
|
|
Term
If one of the independent variables is a lagged value of the dependent variable, ______ will cause an inaccurate parameter estimate. |
|
Definition
|
|
Term
There are too many dummy variables specified, so the equation will suffer from: |
|
Definition
|
|
Term
Simple Linear Regression:
Correlation (formula)
Estimated slope coefficient (formula)
Estimated intercept (formula)
Confidence interval for predicted Y-value |
|
Definition
Correlation:
rxy = covxy / [sxsy]
Estimated slope coefficient:
ESC: covxy / sx2
Estimated intercept:
EI: b(hat)0 = Y(bar) - b(hat)1X(bar)
Confidence interval for predicted Y-value:
Y(hat) +/- [tc*SE(forecast)] |
|
|
Term
Multiple Regression:
Test statistical significance of b; H0=?; #df?; reject if?
Confidence interval |
|
Definition
H0: b=0
t = b(hat) / sb(hat) with n-k-1 df
Reject if |t|>tc or p-value < (alpha)
Confidence Interval:
b(hat)j +/- (tc*sb(hat)) |
|
|
Term
Total Sum of Squares (SST) (formula) (definition) |
|
Definition
SST = RSS + SSE
(Regression sum of squares + sum of squared errors)
Measures the total variation in the dependent variable. Not the same as variance. |
|
|
Term
|
Definition
|
|
Term
Heteroskedasticity (def) (detect with) (correct with) |
|
Definition
Non-constant error variance
Detect with Breusch-Pagan test
Correct with White-corrected standard errors |
|
|
Term
Autocorrelation (def) (detech with) (positive autoc. if) (correct by) |
|
Definition
Correlation among error terms; error terms are not independent
Detect with Durbin-Watson test
Positive autoc. if DW<d1
Correct by adjusting standard errors using Hansen method |
|
|
Term
Multicollinearity (def) (detect by) (correct by) |
|
Definition
High correlation among Xs (independent variables)
Detect if F-test significant + t-tests insignificant
Correct by dropping X variables |
|
|
Term
Linear trend model
Log-linear trend model |
|
Definition
yt = b0 + b1t + et
ln(yt) = b0 + b1t + et |
|
|
Term
Covariance stationary (def) (how to determine if time series is C.S.) |
|
Definition
mean and variance don't change over time
To determine:
(1) plot data
(2) run an AR model and test correlations
(3) perform Dickey Fuller test |
|
|
Term
Unit Root (def+) (how to eliminate U.R.) |
|
Definition
coefficient on lagged dependent variable = 1. Series with UR is not covariance stationary.
First differencing will often eliminate the unit root |
|
|
Term
Autoregressive (AR) model specified correctly if: |
|
Definition
autocorrelation of residuals not significant |
|
|
Term
Seasonality (indicated by) (correct by) |
|
Definition
Indicated by statistically significant lagged err. term
Correct by adding lagged term |
|
|
Term
Autoregressive Conditional Heteroskedasticity (ARCH):
-detected by estimating
-Variance of ARCH series
|
|
Definition
e(hat)t2 = a0 + a1e(hat)2t-1 + (mu)t
Variance:
se(hat)2t+1 = a(hat)0 + a(hat)1*error(hat)t2 |
|
|
Term
Beta (stock/market and asset/portfolio) |
|
Definition
Beta (stock to market) = COV(stock, market) / Var(Mkt) Beta (asset to portfolio) = COV(asset, portfolio) / Var(Port) |
|
|
Term
|
Definition
covxy = Sigma[[Xi - X(mean)]*[Yi-Y(mean)]]
/
(n-1) |
|
|
Term
The coefficient on each dummy tells us: |
|
Definition
the difference in (whatever) between the respective quarter and the one left out |
|
|
Term
Durbin-Watson decision rule |
|
Definition
H0: r = 0
If DW statistic > upper boundary, FAIL TO REJECT
(no serial corr.)
If DW statistic < lower boundary, REJECT
(there is serial corr.)
If lower bound<DW<upper bound, INCONCLUSIVE
|
|
|