Pelican Parts
Parts Catalog Accessories Catalog How To Articles Tech Forums
Call Pelican Parts at 888-280-7799
Shopping Cart Cart | Project List | Order Status | Help



Go Back   Pelican Parts Forums > Miscellaneous and Off Topic Forums > Off Topic Discussions


Reply
 
LinkBack Thread Tools Rate Thread
Author
Thread Post New Thread    Reply
jyl jyl is offline
Registered User
 
jyl's Avatar
 
Join Date: Jan 2002
Location: Nor California & Pac NW
Posts: 19,905
Garage
Calling Excel Gurus

This is sort of an Excel question.

Here is Census data on household income for Portland OR. Mean income has risen about 5%/yr from 2011-2018.



I am trying to see if the gain in mean income, and the decrease [increase] in the number of low [high] income households, is due to broad across the board income gains, or to high income households migrating in and low income households migrating out.

It occurred to me that one way to test this would be to 1) build an assumed model for the 2010 household income distribution in some detail (X households with $0-1K income, Y households with $1-2K income, all the way to ZZ households with $200K+ income), 2) adjust that model to match the reported 2010 household counts by income ranges and the 2010 mean and median income, then 3) start testing different scenarios. For example, grow every household's income by 5%/yr and assume net in-migration was of the same income distribution as existing households, and see if the resulting modeled population and household distribution match the reported 2018 data. If yes, then it is plausible that incomes did grow broadly across the board. If not, start testing other scenarios to see which might result in the model matching the 2018 data.

Does this make any sense at all? It'll be a lot of work.
__________________
1989 3.2 Carrera coupe
1988 Westy Vanagon, Zetec conversion
Really want to find a clean E28 . . .
Old 10-07-2019, 02:40 PM
  Pelican Parts Catalog | Tech Articles | Promos & Specials    Reply With Quote #1 (permalink)
=VLOOKUP(val,tab,col,rng)
 
RANDY P's Avatar
 
Join Date: Mar 2002
Location: Inside Dpmulvan's head. LOL.
Posts: 19,794
Garage
You accounting for inflation?
Why can't YoY changes in each category for income / headcount work?
__________________
QUOTE: NOTA, on dating: I have a simple system-I ask them to put it in
--if they donot it don't go in
Old 10-07-2019, 03:22 PM
  Pelican Parts Catalog | Tech Articles | Promos & Specials    Reply With Quote #2 (permalink)
jyl jyl is offline
Registered User
 
jyl's Avatar
 
Join Date: Jan 2002
Location: Nor California & Pac NW
Posts: 19,905
Garage
This is all nominal.

I tried simply growing all income levels by a uniform 5%/yr, and that results in mean income +5%, not surprising I guess. But I want to test if that also results in a 2018 income distribution that matches the actual.
__________________
1989 3.2 Carrera coupe
1988 Westy Vanagon, Zetec conversion
Really want to find a clean E28 . . .
Old 10-07-2019, 03:28 PM
  Pelican Parts Catalog | Tech Articles | Promos & Specials    Reply With Quote #3 (permalink)
canna change law physics
 
red-beard's Avatar
 
Join Date: Jul 2000
Location: Houston, Tejas
Posts: 38,239
Garage
Simpliest thing is to graph it.

You really need a regression on each variable or an ANOVA to check the relative slope. There are better programs for statistical work than Excel.
__________________
James
The pessimist complains about the wind; the optimist expects it to change; the engineer adjusts the sails.- William Arthur Ward (1921-1994)
Red-beard for President, 2020
Old 10-08-2019, 05:57 AM
  Pelican Parts Catalog | Tech Articles | Promos & Specials    Reply With Quote #4 (permalink)
Back in the saddle again
 
masraum's Avatar
 
Join Date: Oct 2001
Location: Houston, TX
Posts: 31,850
I'm no statistician, but couldn't you just create a graph with the distribution for each year as a line? Then you colud visually compare each line. You could even have excel create a best fit version of each line and compare those. Then if you wanted to estimate/test alternatives, you could create figured based on the formulas of the best fit lines.

It would be reasonable enough for govt work or handgrenades unless you're planning to write a paper and submit it for peer review, I would think.
__________________
Steve
'08 Boxster RS60 Spyder #0099/1960
- never named a car before, but this is Charlotte.
'88 targa SOLD 2004 - gone but not forgotten
Old 10-08-2019, 06:54 AM
  Pelican Parts Catalog | Tech Articles | Promos & Specials    Reply With Quote #5 (permalink)
Reply

Thread Tools
Rate This Thread
Rate This Thread:

 


All times are GMT -8. The time now is 07:21 PM.


 
Powered by vBulletin® Version 3.8.7
Copyright ©2000 - 2019, vBulletin Solutions, Inc.
Search Engine Optimization by vBSEO 3.6.0
Copyright 2018 Pelican Parts, LLC - Posts may be archived for display on the Pelican Parts Website -    DMCA Registered Agent Contact Page
 

DTO Garage Plus vBulletin Plugins by Drive Thru Online, Inc.