Durbin-Watson Test Quiz

<p>0 to 4. (A)</p> Signup and view all the answers

<p>Little evidence of autocorrelation. (D)</p> Signup and view all the answers

<p>$σ_T (ε_t - ε_{t-1})^2$ (D)</p> Signup and view all the answers

<p>There is no autocorrelation. (D)</p> Signup and view all the answers

<p>There is significant autocorrelation. (D)</p> Signup and view all the answers

<p>The mean of disturbances is zero. (C)</p> Signup and view all the answers

<p>The variance of the errors is constant and finite. (D)</p> Signup and view all the answers

<p>The standard errors may be improperly calculated. (D)</p> Signup and view all the answers

<p>Using a method that adjusts for varying error variances. (A)</p> Signup and view all the answers

<p>The values of independent variables are fixed during estimation. (C)</p> Signup and view all the answers

<p>The variance of the residuals changes across the range of independent variables. (B)</p> Signup and view all the answers

<p>By analyzing the residuals from the regression. (C)</p> Signup and view all the answers

<p>Misleading hypothesis test results. (D)</p> Signup and view all the answers

<p>Dropping one of the collinear variables (A)</p> Signup and view all the answers

<p>Presence of multicollinearity (A)</p> Signup and view all the answers

<p>Ramsey’s RESET test (A)</p> Signup and view all the answers

<p>Reject the null hypothesis (D)</p> Signup and view all the answers

<p>It signifies multicollinearity. (D)</p> Signup and view all the answers

<p>Transform correlated variables into a ratio. (B)</p> Signup and view all the answers

<p>It is used to measure multicollinearity. (C)</p> Signup and view all the answers

<p>They can introduce new problems. (A)</p> Signup and view all the answers

<p>Transform the data into logarithms (A)</p> Signup and view all the answers

<p>The properties of normal distributions are well-understood (C)</p> Signup and view all the answers

<p>Both values should equal 0 (D)</p> Signup and view all the answers

<p>Joint normality of residuals' skewness and kurtosis (A)</p> Signup and view all the answers

<p>Presence of extreme residuals (B)</p> Signup and view all the answers

<p>Using logarithms (C)</p> Signup and view all the answers

<p>Consider using methods that do not assume normality (C)</p> Signup and view all the answers

<p>It is the standardized third moment of the residuals (B)</p> Signup and view all the answers

<p>The variances of the disturbances are equal. (A)</p> Signup and view all the answers

<p>It is the ratio of the two residual variances. (D)</p> Signup and view all the answers

<p>The choice of where to split the sample is arbitrary. (C)</p> Signup and view all the answers

<p>It makes few assumptions about the form of heteroscedasticity. (A)</p> Signup and view all the answers

<p>Coefficients of other variables will be biased and inconsistent. (A)</p> Signup and view all the answers

<p>The dependent variable from the original regression. (D)</p> Signup and view all the answers

<p>Chi-squared distribution (C)</p> Signup and view all the answers

<p>The estimators remain consistent and unbiased. (B)</p> Signup and view all the answers

<p>To ensure uniform variance in residuals. (B)</p> Signup and view all the answers

<p>To confirm that parameters are constant across the sample period. (B)</p> Signup and view all the answers

<p>Var(𝜀𝑡) = σ2 (C)</p> Signup and view all the answers

<p>It compares the RSS of a restricted regression to an unrestricted regression. (D)</p> Signup and view all the answers

<p>Residual Sum of Squares (B)</p> Signup and view all the answers

<p>Estimate the regression for the whole period. (D)</p> Signup and view all the answers

<p>It is biased if the omitted variable is correlated. (C)</p> Signup and view all the answers

<p>It allows for analysis of changes in parameters over time. (A)</p> Signup and view all the answers

The expected value (mean) of the error terms is zero. This assumption ensures that the errors are not systematically biased in one direction.

The variance of the error terms is constant across all observations and finite. This ensures consistent error variability across the data.

The error terms are uncorrelated with each other. This assumption prevents the errors from influencing each other, ensuring independence.

The explanatory variables (X) are not random and remain fixed in repeated samples. This allows us to treat them as constants when estimating the regression coefficients.

The error terms follow a normal distribution with a mean of zero and a constant variance (𝜎2). This assumption is important for hypothesis testing and confidence interval construction.

A situation where the error terms have a constant variance across all observations. This means that the spread of the errors is consistent regardless of the value of the explanatory variables.

A situation where the error terms do not have a constant variance across all observations. This means that the spread of the errors varies depending on the value of the explanatory variables.

The estimates of the regression coefficients are not accurate, meaning they might be biased and systematically different from the true values.

A statistical method used to detect heteroscedasticity, a violation of the assumption of equal variances in the error terms of a regression model.

A value representing the difference between the actual observed value of a dependent variable and its predicted value from a regression model.

Splitting the data into two subsamples and estimating separate regressions for each subsample.

Estimating the regression model using each subsample obtained from the data, and calculating the variances of the residuals for each subgroup.

A statistical test where the null hypothesis is that the variances of the error terms are equal across the two subgroups.

The ratio of the higher residual variance to the lower residual variance, used to assess heteroscedasticity in the Goldfeld-Quandt test.

The distribution of the GQ test statistic under the assumption of homoscedasticity (equal variances); used to determine the p-value of the test.

The choice of where to split the sample in the GQ test, potentially affecting the outcome of the test.

A pattern in residuals where they alternate above and below the time axis more frequently than if they were random. Suggests a relationship between an error term and its previous one.

Absence of any systematic patterns in the residuals. This is the ideal scenario.

A statistical test used to detect first-order autocorrelation, analyzing the relationship between an error term and its previous one.

A statistical value calculated to assess the presence of autocorrelation. A value near 2 indicates no autocorrelation. Values significantly different from 2 suggest the presence of autocorrelation.

The null hypothesis tested by DW test, suggesting that there's no autocorrelation present. We aim to reject this hypothesis if sufficient evidence exists.

The alternative hypothesis in the DW test, suggesting that there is autocorrelation present. We aim to accept this hypothesis if the null hypothesis is rejected.

The condition where the explanatory variables are not random and stay fixed in repeated samples. This allows us to treat them as constants while estimating regression coefficients.

The condition where the regression equation includes a constant term. It's often represented by the 'intercept' in linear regressions.

A transformation applied to data to convert multiplicative relationships into additive ones, simplifying analysis.

A test designed to assess if the residuals in a regression model deviate from a normal distribution. It examines skewness and kurtosis.

The standardized third moment of a probability distribution, measuring the asymmetry of the data.

The standardized fourth moment of a probability distribution, measuring the 'tailedness' or 'peakedness' of the data.

The Bera-Jarque test statistic, calculated based on skewness and kurtosis coefficients, used to assess normality of the data.

The Bera-Jarque test compares the calculated test statistic to a chi-squared distribution with 2 degrees of freedom to determine the significance of the deviation from normality.

A situation where influential data points distort the regression model's fit and can significantly impact the results. Often identified as outliers.

The process of using dummy variables to account for unusual patterns in data, such as outliers, identified in the residuals, helps adjust the model's fit.

A situation where two or more independent variables in a regression model are highly correlated. This can make it difficult to isolate the individual effects of each variable, affecting the accuracy and reliability of the regression results.

When looking at the correlation matrix of variables, you notice a strong correlation between pairs of variables.

A statistical method that uses the variance of the regression coefficient to measure the amount of multicollinearity in a regression model.

A technique to reduce or eliminate multicollinearity by dropping one of the highly correlated variables from the regression model.

A way to handle multicollinearity by creating a new variable that represents the ratio of two highly correlated variables. This can help to capture the combined effect of both variables.

A statistical test used to check if the chosen functional form of a regression model is appropriate. It essentially tests for the presence of misspecification of functional forms.

A test used to check for the presence of heteroscedasticity. It is usually done by regressing the squared residuals from the original regression onto the explanatory variables.

A technique used to create a new variable that takes on a value of 1 for a specific time period and 0 otherwise. This allows you to isolate the effect of a particular observation or time period on the regression results.

The estimated coefficients of included variables in a regression model can be biased and inconsistent if an important variable is omitted, unless the omitted variable is uncorrelated with all the included variables.

Including a variable that has no real relationship with the dependent variable in your regression model does not affect the consistency and unbiasedness of coefficient estimates, but it makes the estimation less efficient.

A statistical test used to determine if the regression coefficients are constant across different time periods. It compares the sum of squared residuals from regressions estimated separately for different time periods to a regression estimated over the entire period.

A type of parameter stability test that assesses whether the estimated coefficients are consistent across different time periods.