 # Some perspective. Dow from 1930. Yearly. Log scale.

4 replies to this topic

### #1 milantrader

Member

• • 511 posts

Posted 07 April 2018 - 06:57 AM

### #2 MaryAM

MaryAM

Member

• • 1,165 posts

Posted 07 April 2018 - 01:10 PM

The DOW formula does not have a constant - making it an invalid mathematical formula much less an average.  It becomes a separate and unique equation every time they replace a stock and/or split and reduce or increase the divisor.  You can only look at a separate chart for each new formula.   If you force a constant into the equation giving it a slope of 1 you get a damped harmonic oscillator - multiply the reported average by the divisor will do the trick.   You simply can't log transform the average number reported over the decades.  Log transformations require that the underlying equation must contain a constant in the formula.   A divisor <1 makes the formula a multiplier - not an average.  Right now 100 points is roughly \$15.

### #3 Dex

Dex

Member

• • 1,919 posts

Posted 07 April 2018 - 02:51 PM

I understand the concept of a log scale, but I think it doesn't add clarity.

"The secret of life is honesty and fair dealing. If you can fake that, you've got it made. "
17_16

### #4 milantrader

Member

• • 511 posts

Posted 09 April 2018 - 04:28 AM

Clarity about what, Dex? You can use the log scale to calculate an underlying trend. I did one from 1900. Not posted yet.The pitchforks (1973 and 2007) are approximately in line with each other. Thats all.

### #5 Dex

Dex

Member

• • 1,919 posts

Posted 09 April 2018 - 02:44 PM

Clarity about what, Dex? You can use the log scale to calculate an underlying trend. I did one from 1900. Not posted yet.The pitchforks (1973 and 2007) are approximately in line with each other. Thats all.

Log scale chart are used when there is large data variances and percentage changes.

Do you get the same results when you use arithmetic scale?

"The secret of life is honesty and fair dealing. If you can fake that, you've got it made. "
17_16