Heck, I dunno. I do graphs with years on the horizontal axis and debt on the vertical.
I wanted to do a graph showing the relation between debt-per-dollar and Shiller's interest rate. Sort of like the Phillips Curve graphs that put inflation on one axis and unemployment on the other. No axis has years. I had to organize the data and create the graph a few times before I got a decent result.
Only decent. I was hoping to get pop-ups that identify the year of each point on the graph. Couldn't get that to work.
I did manage to break up the dots into color-coded chunks by date, as you can see.
| Graph #1 |
Thing of it is, I don't know what to do with that graph. I have more confidence in my years-on-the-x-axis graphs. Like this:
| Graph #2 |
But maybe something will turn up.
Art
ReplyDeleteJust so you know. Nick Rowe has been having problems with his comment section. You may have been caught in a spam filter. He actaully did a post on it a couple days ago. You might try again.
Ill say that Nick has a pretty thick skin and doesnt take comments down very often. Yours probably got lost.
I'd say that both graphs strongy suggest tht there is no relationship between debt/dollar and interest rates.
ReplyDeleteThat, in itself, might be worth knowing.
Cheers!
JzB
I would expect that the relationship would be between the rate of change of debt per dollar and interest rates.
ReplyDeletei.e. something like: "when rates are low, debt is easier to make and so it accumulates faster." So maybe look at that instead?
I guess that I would do yearly percentage growth in DPD (maybe averaged over some small period of time (3 years? don't know) just to make it less noisy).
Interest rates peak first; DPD peaks later. Twice, that I know for sure.
ReplyDeleteInterest rates peak first; DPD peaks later. Twice, that I know for sure.
ReplyDeleteSo if there is any cause and effect, Interest rates are the driver. But with lags >> a decade, that will be a hard case to make.
Cheers!
JzB
It worked. Thanks, Greg!
ReplyDelete