I was a guest on the Jay Thomas Show on WDAY radio out of Fargo today. The guest host was Rob Port from the SayAnythingBlog, Forum Communications Op-Ed pages, TV appearances, and probably a bunch of stuff I am forgetting. He might be getting close to the title of “King of All North Dakota Media” at this point. The topic was forecasting. Yes I know. Friday afternoon in the summer and we were talking forecasting. I do not apologize for it, since I am pretty much always thinking about statistical models.
Several former students asked questions about forecasting performance and I assume it has to do with my recent post about the performance of the forecast of North Dakota sales tax. As they well know and a quick investigation would make clear, there is no shortage of discussion of these issues. The student know this, it is one of the things I tell them is constantly updated in the field.
Sales tax is by far the largest consistent component in revenue generation for the state of North Dakota, but it is by no means the only one. With that in mind I am generating similar information for other revenue streams. Today it will be corporate income taxes. This one is a bit tricky to look at because there are instances of missing values. These are not missing in the sense that somebody is hiding something; these are missing values in the sense that the report provides nothing for the forecast value for corporate income tax in July, August, and September in that particular publication by the state OMB.
The radio discussion today focused on the importance of forecasts and the forecast process in decision-making, particularly budgeting. Most of this related to various aspects of the forecasting and budgeting process at the state level in North Dakota and how it impacts other entities like UND. There was also some appropriate trashing of the forecasters that gave such erroneous predictions and then decide the state is in recession. I discussed some data which I want to make available here. There is more that will be put online as I get it created.
We had an active discussion in the last week regarding consumer opinion/sentiment and spending, the actual amounts they buy. Colleagues were surprised by my assertion that the relationship is not as well defined as many believe. Specifically, improvements in sentiment are essentially meaningless. They do not necessarily translate into changes in spending. The graph here shows the two.