Hacker News Reader

Comment by lode | Hacker News Reader

The Dow is a ridiculous stock index. It doesn't adjust for inflation, and - more importantly - just looks at the stock price, not the underlying market cap. So if Caterpillar (market cap 49.44B) rises 5$ from 80 to 85, the Dow rises 32 points. If Exxon Mobil (market cap 362 Billion) rises from 86 to 91, the Dow rises the same 32 points, even though the first rise means CAT grew only in 3,3 billion Market cap, while XOM grew 20,95 Billion.

NPR's Planet Money has a great episode on the DJI: http://www.npr.org/blogs/money/2013/03/12/174139347/episode-...

View on HN

Agreed, but given their differences, the historical correlation between the S&P and the dow is absurd. https://www.google.com/finance?q=INDEXDJX%3A.DJI&ei=Wef5VPmw...

Fixed link:

https://www.google.com/finance?q=INDEXDJX%3A.DJI%2C+INDEXSP%...

are index linked funds linked to other indices?

Usually S&P500

i suspected that. do you think that would explain the absurd correlation?

The Dow is price-weighted because it was started before computers. Add 30 prices and divide by adjustment factor is the easiest to calculate. Also, it was the first index, before people realized other methods are more accurate.

Now it has to stay that way for backward compatibility.

> Add 30 prices and divide by adjustment factor is the easiest to calculate.

Multiply 30 prices by shares outstanding, and add, is not substantially harder computation to do once a day.

> Now it has to stay that way for backward compatibility

Changing ATT to AAPL isn't backward compatible.

What compatibility is there to maintain? Nothing important depends on the value of the DJIA from before last week.

It is backward-compatible, because they adjust the divisor to compensate, so that the index value pre-rebalancing is the same as the index value post-rebalancing.

If they never adjusted the list of stocks, it would eventually become irrelevant as stocks go bankrupt or are acquired. Only 1 current Dow member was in the index in 1907 (GE), and 4 current Dow members were in the index before 1939 (XOM, PG, DD).

Agreed - a price-weighted index like the DJIA doesn't make much sense for describing the aggregate behaviour of the constituents, even if it does attempt to adjust for stock splits.

A market-cap weighted index like the S&P 500 is probably a better measure.

I don't know much about this, but I looked them up and it's interesting how well the two compare: https://www.google.com/finance?q=INDEXSP%3A.INX%2C+INDEXDJX%...

Yes, the calculation of the Dow Jones is a complete joke, but, funnily enough, it has tracked the fairly sensibly calculated S&P 500 surprisingly closely over time.

No, it hasn't. The two diverge greatly over long windows.

The correlation between SPX and the DJIA will generally be quite high, because sampling 30 stocks (and generally those with the biggest market caps) gives you a good estimate/representation.

As a result, over short periods of time, the spread between the two will be relatively small.

Over long periods of time however, the spread can be significant. See the following:

1. http://avondaleam.com/dow-jones-vs-sp/

2. http://www.thumbcharts.com/101035/DJIA-vs-S-P-500 (Compare 1, 2, 3 and 5 years)

FYI, the S&P500 is weighted by float (market cap available for public trading) and not market cap, per se.

I can't upvote this enough. Once I learned about it, it really shocked me about how this is one of the most popular indexes in the world.

Its popular in the media - but it isn't really used as a benchmark. DIA, the Dow Jones Industrial Average ETF has $12bn AUM whereas SPY, the market-cap weighted S&P 500 etf has $189bn AUM... and probably significantly more if you look at all the "passive" money...

if you buy caterpillar at $80 and sell at $85 you've made 6.25% gain, whereas if you buy exxon mobil at $86 and sell at $91 you've made 5.81% gain.

Even though one is a change (to run with your numbers) of $3.3 billion and the other of $20.95 billion. it's just how shares work.

This is true regardless of how many shares you bought, and regardless of how many shares were outstanding[1] or the company's market caps.

[1] obv other than new issues/splits/etc.

Uh assuming you are serious the parent poster obviously understands this. The point is that it is absurd to have an index where if the total market cap of the tracked companies goes down, the index can go up. This has nothing to do with the returns of buying or selling any particular stock.

One way of looking at the DJIA - If you place $1000 into each of the companies underlying it, the DJIA will reflect your gain/loss in investment. If you tried to weight the index based on the underlying market cap of each of the companies, you wouldn't get a correct assessment of the value of your investment.

Reply on news.ycombinator.com