Question

1. Which one of the following is a good candidate to forecast the cyclical component for the future?
   1. HES
   2. SES
   3. WES
   4. All of the above

2. In OLS, deviations of predicted values from actual values are called
   1. Residuals
   2. Population errors
   3. Random deviations
   4. All of the above

3. When computing the MAt, _________ is(are) removed
   1. Seasonality and irregular fluctuations
   2. Seasonality, irregular fluctuations, and cyclical movements
   3. Seasonality and cyclical movements
   4. Irregular fluctuations and cyclical movements

4. A common source of unusual coefficient estimate signs and statistical significance rulings is
   1. A small sample
   2. An unspecified model
   3. An overspecified model
   4. All of the above

5. If there is a serial correlation in the errors
   1. The standard errors will be smaller than normal
   2. The null hypothesis will be rejected more than usual
   3. Coefficient estimates will be ruled statistically significant more often than normal
   4. All of the above

1. If you have quarterly data and suspect seasonality on a quarterly basis, a ________ is appropriate.
   1. MA(4)
   2. MA(3)
   3. MA(2)
   4. MA(1)

2. Out of the 4 components for many business time-series, which one is the most challenging to forecast?
   1. The cyclical component
   2. The trend component
   3. The irregular component
   4. The seasonal component

3. In trying to forecast the future cyclical component of a time-series, which one of the following might be the most useful type of indicators?
   1. Leading economic indicators
   2. Coincident economic indicators
   3. Lagging economic indicators
   4. Converging economic indicators

4. Time-series decomposition models remain popular today due to the following reasons, except:
   1. They take little computational time to implement
   2. They provide excellent forecasts
   3. They are relatively easy to understand and to explain
   4. They are consistent to the way managers look at data

1. In time-series decomposition, all seasonal indices for the first month for an extended time period are expected:
   1. To be equal
   2. To be similar, but not equal
   3. To fluctuate between 0 and 1
   4. To fluctuate between -1 and 1

2. By design, a dummy variable can take on values
   1. Of 1 or 0 only
   2. Between 0 and 1
   3. Between 0 and 100
   4. Between 0 and infinity (all positive)

3. Which one of the following is a common source of serial correlation in the errors?
   1. Long-term cycles
   2. Non-linear trends
   3. A mispecified model
   4. All of the above

4. As a rule of thumb, if the DW statistic is ______, it’s considered a sign of positive serial correlation in the errors”
   1. <1.5
   2. >=1.5
   3. >=1.5 and <=2.5
   4. >=2.5

5. In Multiple Regression, the y-intercept estimate
   1. Is always positive if the correlation between X and Y is positive
   2. Is always negative if the correlation between X and Y is positive
   3. Can be positive or negative, depending on the correlation between X and Y
   4. Rarely has a useful interpretation

6. Using two independent variables that are highly correlated might lead to
   1. Multicollinearity problem
   2. Heteroskedastic errors
   3. Serially correlated errors
   4. All of the above

7. A violation of the OLS assumption that population errors are uncorrelated and have constant variance is known as
   1. Homoscedasticity
   2. Heteroscedasticity
   3. Kurtosis
   4. All of the above

8. For quarterly data, the MAt will include ____ values before t and _____ values after t.
   1. 2;1
   2. 2;2
   3. 1;2
   4. 4;0

1. In regression analysis, an extrinsic variable is a variable that
   1. Affects endowments
   2. Preferences
   3. Technology
   4. None of the above

2. The cyclical component of a time-series is hard to predict because it doesn’t have a constant
   1. Mean and Variance
   2. Amplitude
   3. Periodicity
   4. All of the above

1. If you have monthly data, the CMAt will have ______ missing values
   1. 6
   2. 5
   3. 12
   4. 11

2. The Alkaline Information Criteria is said to be based on the parsimony principle because
   1. It tends to be better than the Bayesian Information Criteria for large enough samples
   2. The statistic increases as new independent variables are added to the model
   3. It requires a large enough sample
   4. It is very simple to compute

3. Multicollinearity can be easily identified when using
   1. A correlation matrix for the independent variables
   2. A correlation statistic between the dependent variable and each independent variable
   3. A graph of autocorrelation coefficients
   4. The Theil Statistic

4. In multivariate regression models, the best model
   1. Minimizes the adjusted R-Square
   2. Minimizes the AIC or BIC statistic
   3. Maximizes the AIC or BIC statistic
   4. Maximizes the DW statistic

5. In time-series decomposition, all seasonal factors for the first month for an extended time period are expected
   1. To be similar, but not necessarily equal
   2. To be equal
   3. To fluctuate between 0 and 1
   4. To fluctuate between -1 and 1

6. An approximation of the forecast confidence interval is
   1. y-hat ± 2 (SEE)
   2. y-hat ± 1 (SEE)
   3. y-hat ± 3 (SEE)
   4. y-hat ± 2 (t-stat)

7. By construction, the DW statistic has values between
   1. 0 and 1
   2. Plus one and minus one
   3. Plus 4 and minus 4
   4. 0 and 4

8. If P represents the different states of nature for a variable, you can add at most _____dummy variables in the regression
   1. P-1 variables
   2. P variables
   3. 1 variable
   4. P+1 variables

9. The time-series decomposition forecast is the product of the four components, and the irregular component is assumed to be:
   1. 1, unless the forecaster has a reason to believe that a shock may take place
   2. 0, unless the forecaster has a reason to believe that a shock may take place
   3. 1
   4. 0

10. HES is a good candidate to be used to project the cyclical component into the forecast horizon, except that:
    1. HES cannot predict a turning point until after it has occurred
    2. HES has no way to account for trend in the cyclical component
    3. HES can only be used with historical data and cannot be used for future forecasts
    4. All of the above

Answer 1

desi dumYariable com fake valus Voviomce Mushelineciycan bal e can be easily athen sih er vaa bes 13