Decades charts - SP500, PE10, inflation

[Lsbcal]

Updated charts with a few new ones thrown in. As usual this is historic, this stuff is not predictive. But you can wonder and prognosticate.

Next week there will be a test on what caused all the wiggles.

 

[aja8888]

Those look like some of the 3D seismic charts we see offshore in 2D. No wonder why dry holes are drilled after spending millions.

 

[Animorph]

What is PE10 ranking, versus PE10 itself?

 

[Lsbcal]

What is PE10 ranking, versus PE10 itself?

PE10 ranking is taking the current PE10 and ranking it in percentage terms against all the PE10 monthly values since 1920. The rank for say, Jan 1960, would be measured only for the period 1920 to that date. It is not concerned with future (post 1960) absolute values.

I did not want to look at just the absolute value but rather how the PE10 compared to all past values. Is this a better way to do it?

 

[Onward]

Updated charts with a few new ones thrown in.

Once again, thanks for these charts. Historical perspective is valuable.

 

[Animorph]

With all the PE10 values on a chart it would be easy to compare with the past, and future. This way, I'm not sure something is being hidden in the processing. If a salesman showed me that chart, I'd be suspicious.

 

[Lsbcal]

With all the PE10 values on a chart it would be easy to compare with the past, and future. This way, I'm not sure something is being hidden in the processing. If a salesman showed me that chart, I'd be suspicious.

I couldn't sell anything to save my soul.

 

[aja8888]

With all the PE10 values on a chart it would be easy to compare with the past, and future. This way, I'm not sure something is being hidden in the processing. If a salesman showed me that chart, I'd be suspicious.

I'd be suspicious too, especially if he said "it would be easy to compare with the past and the future".

 

[Lsbcal]

Have you guys ever bought anything from a blue rabbit?

If not you should hurry to get a copy of the latest Lsbcal newsletter with 2014 chart extensions: https://www.google.com/search?q=ima...0302-new-ocean-trash-garbage-patch%2F;600;398

 

[photoguy]

PE10 ranking is taking the current PE10 and ranking it in percentage terms against all the PE10 monthly values since 1920. The rank for say, Jan 1960, would be measured only for the period 1920 to that date. It is not concerned with future (post 1960) absolute values.

I did not want to look at just the absolute value but rather how the PE10 compared to all past values. Is this a better way to do it?

What's the argument for (1) using ranking instead of absolute values and (2) using the data only up to that date? I haven't heard of this before and I'm curious as to the rationale.

 

[Lsbcal]

What's the argument for (1) using ranking instead of absolute values and (2) using the data only up to that date? I haven't heard of this before and I'm curious as to the rationale.

I use ranking like this as only one metric with others to manage my investments. The logic then behind using data only up to the date is that the use of that data at that point was at that time the only data set available. If one were making decisions based on this data, that would be their end point. As of Jan 2014 we can look back at the full data set to rank (but not forward unfortunately). Actually this technique was used by Wade Phau in another PE10 context.

One can just ignore my ranking graph and look at the other graph I posted above which is a more traditional way of viewing the data. Note that graph has some useful info like the standard deviation of 10 year earnings.

 

[NW-Bound]

This chart clearly shows that the decade of 2000-2009 was the most terrible in the last 70 years. It was even worse than 1970-1979. Son of a gun!

And most of us here survived that. I even prospered and doubled my NW and portfolio too, but that was because I still had part-time income, which I do not anymore. Without that, I would be in trouble I think, because of expenses with my children education. I would also not have the courage to [-]buy low/sell high[/-] rebalance during the two terrible crashes.

I recall some posters who lamented personal financial tightness during 2008-2009, then disappeared altogether. Here's a toast to all of us who still stick around, however we did it. Let's hope that we will also survive whatever comes next.

PS. The chart also shows how we could retire early. Two consecutive decades of stock growth, and bond too, of 1980-2000 compounded together were the driver. We were lucky more than smart.

PPS. Prior to us, our elders also enjoyed two consecutive decades of growth, i.e. 1950-1969. They then suffered the bad decade of 1970-1979. Note that the prosperity of 1950-1069 did not quite match the boom years of 1980-2000, but their bad decade was also more benign.

 

[photoguy]

The logic then behind using data only up to the date is that the use of that data at that point was at that time the only data set available. If one were making decisions based on this data, that would be their end point.

Thanks for the explanation. Using the rank up to that date would not be my preference for visualization but I totally understand the need for it when testing a decision rule.

 

[Fred123]

Hmmm, not so sure that this is the best way to visualize this data since it's very sensitive to the endpoints you choose. For example, if you chose 1997 as the start of one of the decadal S&P 500 time series instead of 2000, you'd come away with a completely different impression. Wouldn't a 10 year moving average be better?

 

[Finder]

Hmmm, not so sure that this is the best way to visualize this data since it's very sensitive to the endpoints you choose. For example, if you chose 1997 as the start of one of the decadal S&P 500 time series instead of 2000, you'd come away with a completely different impression. Wouldn't a 10 year moving average be better?

First of all, thank you Lsbcal for sharing this work and info. It is helpful.

Regarding the question of rolling PE10, isn't that what is available at:

Shiller PE Ratio

 

[Lsbcal]

Hmmm, not so sure that this is the best way to visualize this data since it's very sensitive to the endpoints you choose. For example, if you chose 1997 as the start of one of the decadal S&P 500 time series instead of 2000, you'd come away with a completely different impression. Wouldn't a 10 year moving average be better?

This is no more then taking a continuous series, segmenting it arbitrarily and shifting each segment to the origin. One could always try visualizing reconnecting some of these segments, like the last 36 months of the 1990's (yellow) to the 2000's (orange) and the current 2010's (blue). The slopes don't change and the y-axis variations stay the same.

M* allows one to easily plot 60 month rolling averages. Apparently not 10 year. But then you loose the sense of volatility that was experienced.

So I think this chart has its uses but is not perfect.