Beware Of Sophistry About The Yen And Nikkei

|
Includes: EWJ, FXY
by: Marc Chandler

Summary

Showing two times series in a chart with two different scales is misleading.

There is a proper (robust) way to look at the two time series and that is to index them.

One can also conduct correlation exercises--correlation can not be eyeballed.

There is a common ploy used by many analysts and reporters that often simply does not stand up to close scrutiny, and would in fact be mocked in the university. The ploy is to take two time series and put them on the same chart but use different scales.

Such a ploy often is used to demonstrate a closer relationship between the two variables that is actually the case. A current example is the a chart of the dollar-yen rate and Japanese stocks.

Here is a Great Graphic that was in the Financial Times. It was created from a data base provided by Thomson Reuters. The chart begins at the start of 2013. It seems like a logical place to begin as Abe was elected at the end of 2012. However, what makes the 10,000 level of the Nikkei (left scale) be at the same level as JPY80 (right scale)?

The truth is there is no compelling reason except it makes the chart more aesthetically pleasing and allows one to argue the closeness of the fit between the yen and Nikkei. Investing is many things, but is most certainly is not a beauty contest.

Click to enlargeHere is the same to time series indexed so that each starts at 100 at the start of 2013. It was created on Bloomberg. It shows that the relationship is not nearly as tight as is suggested by the first chart. This is important for investors. The relationship can inform investment strategies and hedging strategies.

The first chart looks as if the Nikkei and yen have performed similarly. The lower chart demonstrates that this is not the case. Beware.

Implicit too in the sophistry is the idea that correlations can be eyeballed. Perhaps sometimes they can, but it reminds me of when I recently went canoeing. I stuck the oar in the water and looked down. It looked like the oar was bent. I quickly pulled the oar out of the water and found it was still straight and true.

The correlation between the Nikkei and dollar yen moves around quite a bit, which is something else that the first chart does not seem to reflect sufficiently. The correlation between the percentage change in the Nikkei and the percentage change in the dollar-yen rate over the past 100 sessions is near 0.40. Last September is was at the highest level of the Abe's tenure near 0.63. Last June, this 100-day correlation (on the percentage change) dipped into negative territory.

It is said that Euclid once claimed that there was not royal road to geometry, which meant there was no shortcut. In my experience, the same is true about investment. There is no substitute for rigorous critical thinking, and doing one's homework. Be suspicious of fallacious thinking regardless of source or form.

Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.

I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it. I have no business relationship with any company whose stock is mentioned in this article.