Okay, I think I have a fix on this. Suppose we want to test our hypothesis that otherwise comparable firms with high levels of indebtedness cut back on expenditures desired by employees in circumstances where their not-so-indebted cousins would not. (For a further explanation of that hypothesis, review Part I of this discussion from last week.) How do we do it? We'll assume that we haven't found a smoking gun memo in which the company's Treasurer writes to the CEO and says, "we can't afford those darn safety vests any longer. Tell Human Resources to stop buying them so we can make the interest payments!" Assume we're looking at circumstantial evidence. What counts as evidence? What we can't do is simply say: firm X buys safety vests for its employees and is mostly equity financed. Firm Y doesn't and isn't. No matter how many Xs and Ys we find compliant with our hypothesis, we will still have only correlation, not causation. The arrow of...
Posts
My recent reading includes a review, by friend Henry, of SLAVISH SHORE, a new book from Harvard University Press about the life of Richard Henry Dana Jr. Dana argued the Prize Cases before the US Supreme Court in February 1863. At issue: when is a war a war? when is a blockade a blockade? was Lincoln's action in pursuing a war through the imposition of a blockade lawful? As Henry explains it: "A blockade is a war measure, and a President is authorized under his war powers to impose a blockade only when Congress has declared war against another nation....Lincoln denied that the Confederate states were a nation, he insisted that they were mere insurrectionists and traitors. No one questions that Lincoln could have closed the ports, as opposed to blockading them, but Britain had 'made it clear that it would not accept American "closure" of Southern ports that would expose British shippers to arrest as common smugglers.'" In that last half sentence, ...
I'm looking again at the March issue of Harper's . It includes a story by Lesley Nneka Arimah, about a Nigerian immigrant to the United States named Glory and her relationship with a man, Thomas. The Nigerian-immigrant community in the US is, I gather, fairly traditional in its views on marriage, sex roles, etc., and Arimah's story plays on that. I wouldn't call Glory an anti-hero exactly, but the narrative voice's view of the protagonist is, at best, ambivalent. We're told, "It was also true that her parents put pressure on her. Yet theirs was the sort of hopeful pressure that would have encouraged a better person." In time, the parental pressure comes to a point on this: Glory should find a nice Nigerian man and settle down. That leads to the following neat passage: "But the more he said 'us' and 'we,' the less quickly she deleted that 'Mom, I'm seeing someone' text. One day, instead of sending it, she poste...
I just want to record here a statistical fact so I won't forget it down the road. I wrote a story recently for AllAboutAlpha concerning the different prices of "Brent crude" and WTI crude. Brent crude is sold through ICE Futures Europe: WTI through Nymex. Incidental to that story, I also referred to the fact that quantities of Brent crude oil are defined in "cargoes." I didn't answer the question: how many barrels in a "cargo"? Anyway, I'll do so here. Cargo size [600,000 barrels (95,000 cubic meter)] Related facts: The ICE Futures Europe symbol for Brent crude futures is B, also traded on the NYMEX, with the symbol BZ. It was originally traded on the open outcry International Petroleum Exchange in London, but since 2005 has been traded on the electronic Intercontinental Exchange, known as ICE. One contract equals 1,000 barrels (159 m3). Contracts are quoted in U.S. dollars
How can we use statistics to get at causal impact? Of course, correlation does not imply causation. Remember my lesson from a Dilbert's comic last week. But still, there are inferences to be drawn somehow. It may be a fact that MacDonald's just spent a lot of money on an ad campaign involving firecrackers and tapdancing penguins. It may be, also, that they sold a million hamburgers worldwide yesterday. It does not follow that they sold the million hamburgers because of that ad campaign. What MacDonald's clearly wants is some idea of how many hamburgers they would have sold if they had had some other ad campaign in place, possibly some less expensive campaign. Something involving their familiar clown and his sidekicks. Here's a social sciency thing about "Bayesian structural time-series models" that aims at getting at the problem. It's actually based on the experiences of Google, paid clicks versus free clicks, etc. -- I just prefer the MacDonald...
Trying to imagine the scene over in Clinton headquarters when they first saw certain footage of a Sanders rally on March 26th. "Really? Cute birds hopping onto his podium now? so he can sermonize about world peace? We're supposed to run against Francis of Assisi?" Given their penchant for co-opting Bernie, they'll be hiring a bird trainer any minute now. Happy Easter everyone. I had nothing prepared for a proper Easter post, but the St. Francis style moment at least gives us a good religious/historical analogy for the day. Thanks for that Bernie. I hope you had a good Purim.
One of the Dilbert cartoons that came up on my day-to-day desk calendar recently offers a valuable lesson. Correlation does not imply causation. Usually, the pointy-haired boss plays the stooge in his exchanges with Dilbert, but in this case its the boss who brings the point home. Dilbert tells him, "Studies show that companies with a high level of trust in employees also perform the best." Dilbert is presumably hoping that the boss will infer from this correlation of performance and trust that there is a causative link, and that the boss ought to trust Dilbert more, thereby allowing their common employer to reap the rewards. Instead, the boss acutely responds, "If you ever start performing well, I'll trust you, too." Ah, yes, one of the problems with that sort of causal argument (but only one of them) is that the causal arrow is easily reversed. Good one, boss.