5 ms·
> Not all your audience are Cannucks - Your data is beautifully indexed but only related to CAD. It would be helpful if a few pointers to some other well known
by jbay808 4y ago
> Not all your audience are Cannucks - Your data is beautifully indexed but only related to CAD. It would be helpful if a few pointers to some other well known currencies was provided for scale. I too have this snag quite often being a snivelling GBP user!
Thank you! That's a good point. I'm not sure how to handle it because exchange rates do fluctuate on the scale of decades that my post covers. For example in 2012, the CAD and USD were at par (1 USD = 1 CAD), whereas last year we were down as low as US 0.70 per CAD. If I expressed the Canadian CPI in USD that would look like a 30% deflation. Now we're up at 0.79 USD per CAD one year later. It's pretty volatile.
> I note that the graphics of a biplane looks suspiciously like a German WW1 aircraft - the tailfin shape is certainly not a Sopwith Camel or Pup!
You are (almost) correct! It's a Fokker D.VIII monoplane. I'm not a spy! I was just looking for a nice clean public domain vector art image of a prop plane in profile on Wikimedia Commons and that's what I found first. Honest!
- Majromax 4y ago> If I expressed the Canadian CPI in USD that would look like a 30% deflation. Now we're up at 0.79 USD per CAD one year later. It's pretty volatile. If you wanted to convert the numbers, over decadal timescales you'd probably want to use purchasing power parity (PPP, or $INT) figures. That tends to be less volatile, and since it reflects the approximate cost of goods and services it's a better intuitive like-for-like match. > I was just looking for a nice clean public domain vector art image of a prop plane Given the topic, I'm a bit disappointed you couldn't get an image of a Japanese Zero. Incidentally, if you're still working on "Inflation, part 2", one way to close the central bank reaction function is to assume a Taylor Rule (https://en.wikipedia.org/wiki/Taylor_rule https://en.wikipedia.org/wiki/Taylor_rule). I've long thought that the Taylor Rule should be interpreted as a kind of PID controller, save that nobody in economics seems to relate the "assumed equilibrium real interest rate" to an integral term.
- jbay808 4y ago> Japanese Zero Brilliant. I could have made the confusion so much worse. > I've long thought that the Taylor Rule should be interpreted as a kind of PID controller... Yes, this exactly. I'm writing a part 2 from exactly that perspective, and I also want to examine how forward guidance reduces phase margin. But there are a few more concepts I have to lay the groundwork for beforehand, especially how to model forward-looking expectations. It might not be ready particularly soon... I wasn't expecting so much attention so quickly!
- gerdesj 4y ago"You are (almost) correct! It's a Fokker D.VIII monoplane." My eyes are dim, my back is grey etc! Of course it is - I made a model of it about forty years ago and obviously a 1/32 scale Fokker DRI too, painted red with a white nacelle. My parents were stationed in West Germany for about 10 years on and off. The world was quite odd in the 1980s as it is now. Now as to scaling currencies. I'm a bit ambivalent here. I think a rule of thumb might be better than nothing so 4DM == £1! No, as you say it is too volatile for that. I can remember the days when there were at least two USD to the GBP and the Euro wasn't invented. The NLG (Dutch Guilder) was roughly like a DM (Deutsche Mark) but not quite: the Dutch currency was extremely colourful and way prettier than other currencies. France had their Franc - 10 to the quid (GBP) which was easy to convert, and so on. If I want to scale your currency to mine or another then I think that is my problem, now I've had a think about it.
- jbay808 4y agoThanks! Anyway, I'll try to think about ways to help my writing be more approachable for an international audience in my next articles. And regardless, I really enjoyed your comments. Cheers!