Question

Suppose you collect a time series of data and estimate a regression line. Afterwards, you calculate the residuals, which are shown in the following spreadsheet: residuals Plot the residuals over time. Based on the graph, there appears to be [Select ] trend in the residuals over time. Question 18 0.27 pts Based on the previous question, you decide to test for autocorrelation among the residuals. What is the Durbin-Watson statistic? Round to two decimal places, if necessary. Question 19 0.52 pts What are the critical values to compare with the previous Durbin-Watson statistic? Assume a significance level of a = .01. Use this table and assume we only had one independent variable. d = du = Question 20 0.27 pts Based on the answers to the previous two questions, what do you conclude? There is [Select ] evidence of [Select ] autocorrelation.

Answer 1

The Durbin-Watson statistic is used to test for autocorrelation among the residuals in a regression analysis. It measures the extent to which the residuals are correlated with each other over time. A value close to 2 indicates no autocorrelation.

Step-by-step explanation:

The Durbin-Watson statistic is used to test for autocorrelation among the residuals in a regression analysis. The test statistic measures the extent to which the residuals are correlated with each other over time. It ranges from 0 to 4, where a value close to 2 indicates no autocorrelation, a value less than 2 indicates positive autocorrelation, and a value greater than 2 indicates negative autocorrelation.

To calculate the Durbin-Watson statistic, you divide the sum of the squared differences between adjacent residuals by the sum of the squared residuals. The formula is: Durbin-Watson = sum((residual_t - residual_t-1)^2) / sum(residual^2).