Data source is yahoo finance, tools are python and matplotlib
[Constituent Weights] (as of last monthly rebalance)
Sanmina Corporation: 6.67%
Aflac Incorporated: 6.67%
The Walt Disney Company: 6.67%
Capital One Financial Corporation: 6.67%
Westlake Corporation: 6.67%
Freeport-McMoRan Inc.: 6.67%
Blackstone Inc.: 6.67%
Enterprise Products Partners L.P.: 6.67%
Air Products and Chemicals, Inc.: 6.67%
Watsco, Inc.: 6.67%
Skechers U.S.A., Inc.: 6.67%
Penske Automotive Group, Inc.: 6.67%
Berkshire Hathaway Inc.: 6.67%
BlackRock, Inc.: 6.67%
LVMH Moët Hennessy Louis Vuitton SE: 6.67%
not-picky on
The effect seems most pronounced at exactly the start of the chart. Are we sure we haven’t simply selected a date range that most produces this outcome?
elijha on
I would theorize that you’re flipping the causation here. Being a CEO over 70 doesn’t mean you’ll outperform. Being a CEO over 70 who *doesn’t* outperform means you’ll get put out to pasture by the board.
szakee on
correlation is not causation.
hungarian_conartist on
Do the returns not not index. If you started this plot in the middle of 2021 instead of 2020 it might be the S&P500 outperforming grey matter.
Morisior on
How much of this outperformance survives if you disregard Berkshire Hathaway?
RolandSnowdust on
That time frame is not statistically significant.
lucianw on
Hot take: if you plot a graph like this of several time series and you arbitrarily pick t=0 as the point at which they’re all equal, then you must also plot four other graphs where you pick different dates on which they’re equal.
That way you get more intellectual rigor in determining whether the stories the graphs tell are true, or just a visual artifact of your arbitrary starting point.
Hattix on
Instead of using absolute age as a proxy, use the CEO’s length of service directly.
egoVirus on
„Hello, Correlation? Hi, my name is Causation, have you got a minute?“
Leave A Reply
Du musst angemeldet sein, um einen Kommentar abzugeben.
10 Kommentare
Data source is yahoo finance, tools are python and matplotlib
[Constituent Weights] (as of last monthly rebalance)
Sanmina Corporation: 6.67%
Aflac Incorporated: 6.67%
The Walt Disney Company: 6.67%
Capital One Financial Corporation: 6.67%
Westlake Corporation: 6.67%
Freeport-McMoRan Inc.: 6.67%
Blackstone Inc.: 6.67%
Enterprise Products Partners L.P.: 6.67%
Air Products and Chemicals, Inc.: 6.67%
Watsco, Inc.: 6.67%
Skechers U.S.A., Inc.: 6.67%
Penske Automotive Group, Inc.: 6.67%
Berkshire Hathaway Inc.: 6.67%
BlackRock, Inc.: 6.67%
LVMH Moët Hennessy Louis Vuitton SE: 6.67%
The effect seems most pronounced at exactly the start of the chart. Are we sure we haven’t simply selected a date range that most produces this outcome?
I would theorize that you’re flipping the causation here. Being a CEO over 70 doesn’t mean you’ll outperform. Being a CEO over 70 who *doesn’t* outperform means you’ll get put out to pasture by the board.
correlation is not causation.
Do the returns not not index. If you started this plot in the middle of 2021 instead of 2020 it might be the S&P500 outperforming grey matter.
How much of this outperformance survives if you disregard Berkshire Hathaway?
That time frame is not statistically significant.
Hot take: if you plot a graph like this of several time series and you arbitrarily pick t=0 as the point at which they’re all equal, then you must also plot four other graphs where you pick different dates on which they’re equal.
That way you get more intellectual rigor in determining whether the stories the graphs tell are true, or just a visual artifact of your arbitrary starting point.
Instead of using absolute age as a proxy, use the CEO’s length of service directly.
„Hello, Correlation? Hi, my name is Causation, have you got a minute?“