[EMAIL PROTECTED] (Norm Loomer) wrote in message news:<[EMAIL PROTECTED]>... > A colleague brought me some annual time series data that shows a > downward trend from 1992 to 2001. In 1996 the agency that recorded the > data adopted a new program that, if successful, would cause the data to > trend downward more sharply. What they are looking for, I think, is > evidence that the slope after 1996 is greater (in absolute value) than > the slope before 1996. Can someone help me with this? >
The approach I'd use depends somewhat on the autocorrelation structure (and also on other important explanatory variables, if any) e.g. If the errors had small autocorrelation, I might use regression with an allowance for a trend change (the linear term being the first trend, and an additional variable to pick up the change). if the series was close to a random walk, I'd take first differences and then look at whether the post intervention mean was lower than (more negative) than the pre-intervention mean. (I mean some kind of hypothesis test of the change in population mean) If the autocorrelation was middling, I'd probably use a time series regression model (i.e. use a regression-like model that also accounted for the correlation in the errors). Glen . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
