Box car counts by RR - NMRA Charles collection


Larry Kline
 

I just posted a table in the files section file that compares expected and actual counts of boxcars by railroad in the NMRA Charles collection. The file name is _Compare box car counts US RRs - Charles collection.pdf_.

The first two columns in the table list, in descending order, the number of box cars owned on 12-31-1950 by 71 US railroads. These numbers are from _The Handbook of American Railroads_.

There are 194 photos of boxcars in the Charles collection. Most of them were taken in 1946 and 1947 in the greater Harrisburg PA area on the PRR. The most common photo location is the PRR eastbound receiving yard in Harrisburg. The third column in the table lists the expected boxcar counts by railroad for a total of 194 boxcars, based on the second column. The fourth column lists the actual number of boxcar photos in the Charles collection.

Given the small size of the sample, I think that these numbers support Tim Gilbert's hypothesis that the number of boxcars owned by railroad can be used to predict the expected number of boxcars observed. Other list members are welcome to draw their own conclusions about this data.

Larry Kline
Pittsburgh, PA


Rob Kirkham <rdkirkham@...>
 

Hi Larry,

Its an interesting document both for how it correlates and how it does not. I don't suppose you have another version that incorporates the (controversial?) Canadian railway equipment? I wonder how that would impact the figures.

Rob Kirkham

----- Original Message -----
From: "Larry Kline" <lndkline@verizon.net>
To: "STMFC list" <STMFC@yahoogroups.com>
Sent: Saturday, August 16, 2008 7:44 AM
Subject: [STMFC] Box car counts by RR - NMRA Charles collection


I just posted a table in the files section file that compares expected
and actual counts of boxcars by railroad in the NMRA Charles
collection. The file name is _Compare box car counts US RRs - Charles
collection.pdf_.

The first two columns in the table list, in descending order, the
number of box cars owned on 12-31-1950 by 71 US railroads. These
numbers are from _The Handbook of American Railroads_.

There are 194 photos of boxcars in the Charles collection. Most of
them were taken in 1946 and 1947 in the greater Harrisburg PA area on
the PRR. The most common photo location is the PRR eastbound receiving
yard in Harrisburg. The third column in the table lists the expected
boxcar counts by railroad for a total of 194 boxcars, based on the
second column. The fourth column lists the actual number of boxcar
photos in the Charles collection.

Given the small size of the sample, I think that these numbers support
Tim Gilbert's hypothesis that the number of boxcars owned by railroad
can be used to predict the expected number of boxcars observed. Other
list members are welcome to draw their own conclusions about this data.

Larry Kline
Pittsburgh, PA

------------------------------------

Yahoo! Groups Links





--
No virus found in this incoming message.
Checked by AVG.
Version: 7.5.526 / Virus Database: 270.6.3/1614 - Release Date: 8/15/2008 5:29 PM


Larry Kline
 

Rob Kirkham wrote:
Its an interesting document both for how it correlates and how it does
not. I don't suppose you have another version that incorporates the
(controversial?) Canadian railway equipment? I wonder how that would
impact the figures.

There are 4 CP boxcars and 2 CN boxcars in the Charles collection. The
_Handbook of American Railroads_ doesn't include Canadian Railroads. I
added the CN and CP boxcar totals from the April 1951 ORER and revised
the table. Its in the files section with the file name _Compare box
car counts N Amer RRs - Charles collection.pdf_.

Larry Kline
Pittsburgh PA


Rob Kirkham <rdkirkham@...>
 

Thanks Larry! Apart from the PRR, my impression - just scanning through the numbers - is that by incorporating the Canadian numbers (which are obviously underrepresented by fleet size, reflecting the border issues) the deviation between expected and actual in the Charles collection is reduced a little. I suppose one could test for that in Excel, although my first year stats course is more than a 1/4 century ago, so I won't be the one to do it...

A question: what is the Handbook of American Railroads - that's a new one to me. Is it analogous to the ORER? Published how frequently?

Rob Kirkham

----- Original Message -----
From: "Larry Kline" <lndkline@verizon.net>
To: "STMFC list" <STMFC@yahoogroups.com>
Sent: Saturday, August 16, 2008 9:51 AM
Subject: [STMFC] Re: Box car counts by RR - NMRA Charles collection


Rob Kirkham wrote:
Its an interesting document both for how it correlates and how it does
not. I don't suppose you have another version that incorporates the
(controversial?) Canadian railway equipment? I wonder how that would
impact the figures.

There are 4 CP boxcars and 2 CN boxcars in the Charles collection. The
_Handbook of American Railroads_ doesn't include Canadian Railroads. I
added the CN and CP boxcar totals from the April 1951 ORER and revised
the table. Its in the files section with the file name _Compare box
car counts N Amer RRs - Charles collection.pdf_.

Larry Kline
Pittsburgh PA








------------------------------------

Yahoo! Groups Links





--
No virus found in this incoming message.
Checked by AVG.
Version: 7.5.526 / Virus Database: 270.6.3/1614 - Release Date: 8/15/2008 5:29 PM


Dave Nelson
 

-----Original Message-----
From: STMFC@yahoogroups.com [mailto:STMFC@yahoogroups.com] On Behalf Of
Larry Kline

I just posted a table in the files section file that compares expected and
actual counts of boxcars by railroad in the NMRA Charles collection. The
file name is _Compare box car counts US RRs - Charles collection.pdf_.


Larry, there are a couple of errors in that listing and I think you should
correct them. First, as the images were taken on the PRR you should not
count any PRR cars in the sample. Home Road cars are not what the basic
Distribution Hypothesis is about. Second, I get a lower total even after
that so I think you might have missed something when you tabulated the
numbers.

Dave Nelson


Tim O'Connor
 

Larry

I suspect Tim would have recommended computing the standard
deviation of this data set, in order to see how closely these
observed differences from the mean values match the expected
standard deviation. One does not "expect" all values in a
sample data set from a large population to match the mean
values, but one does expect that the mean of the deviations
is predictable. If it is, then you can make a case that your
data set is a good representation of a hypothetical sample.
If not, then the data set may be skewed, or your expected
sample may be incorrect (i.e. Tim's theory of distribution
may be incorrect).

http://en.wikipedia.org/wiki/Standard_deviation

Tim O'Connor

I just posted a table in the files section file that compares expected
and actual counts of boxcars by railroad in the NMRA Charles
collection. The file name is _Compare box car counts US RRs - Charles
collection.pdf_.

The first two columns in the table list, in descending order, the
number of box cars owned on 12-31-1950 by 71 US railroads. These
numbers are from _The Handbook of American Railroads_.

There are 194 photos of boxcars in the Charles collection. Most of
them were taken in 1946 and 1947 in the greater Harrisburg PA area on
the PRR. The most common photo location is the PRR eastbound receiving
yard in Harrisburg. The third column in the table lists the expected
boxcar counts by railroad for a total of 194 boxcars, based on the
second column. The fourth column lists the actual number of boxcar
photos in the Charles collection.

Given the small size of the sample, I think that these numbers support
Tim Gilbert's hypothesis that the number of boxcars owned by railroad
can be used to predict the expected number of boxcars observed. Other
list members are welcome to draw their own conclusions about this data.

Larry Kline
Pittsburgh, PA


Larry Kline
 

Dave Nelson wrote:
Larry, there are a couple of errors in that listing and I think you should correct them. First, as the images were taken on the PRR you should not count any PRR cars in the sample. Home Road cars are not what the basic Distribution Hypothesis is about. Second, I get a lower total even after that so I think you might have missed something when you tabulated the numbers.

I sent my Excel spreadsheet to Dave and we agree on the numbers,

I made a folder in the files section called _Boxcar counts-Charles collection_ and moved the two files I previously posted to the folder. I added a third file with the PRR boxcars omitted.

Larry Kline
Pittsburgh, PA


Larry Kline
 

Tim O'Connor wrote:
Larry, I suspect Tim would have recommended computing the standard deviation of this data set, in order to see how closely these observed differences from the mean values match the expected standard deviation. One does not "expect" all values in a sample data set from a large population to match the mean values, but one does expect that the mean of the deviations is predictable. If it is, then you can make a case that your data set is a good representation of a hypothetical sample. If not, then the data set may be skewed, or your expected sample may be incorrect (i.e. Tim's theory of distribution may be incorrect).

I believe that the random variables in this situation are the boxcar counts for each railroad. The actual boxcar counts for US railroads are not random variables. A single set of data, like the numbers I posted for the Charles collection, has only one value for each random variable so a standard deviation cannot be computed.

On the other hand, if all of the trains in all of the conductor's books owned by list members were treated as separate observations, then standard deviations could be computed for the number of boxcars observed for each railroad. The data in the conductors books could also be aggregated for each railroad.

Larry Kline
Pittsburgh, PA


Malcolm Laughlin <mlaughlinnyc@...>
 

I believe that the random variables in this situation are the boxcar
counts for each railroad.
================

The box car ownership counts themselves are highly biased. A significant percentage of the box cars was our of service. For example, obsolete cars used only in the harvest season for grain loading to a very limited number of destination cities, cars aiting for rebuilding, cars designated for shop programs for special equipment, etc., etc. Depending on the railroad those could account for anywhere from five to 20 percent of the box cars.

There are other biases in the live cars. Like 50' DD or WD cars for lumber loading, cars with six foot doors useful for grain but likely to be rejected by many shippers. There were os many XM box cars with dimensional characteristics that would keep them from being randomly distributed.


Malcolm Laughlin, Editor 617-489-4383
New England Rail Shipper Directories
19 Holden Road, Belmont, MA 02478


boyds1949 <E27ca@...>
 

I don't think we can draw too many conclusions on car distrubution
based on the list of cars in the Charles Collection other than to
marvel at the variety of cars which stopped in Harrisburg. The
selection of cars to photograph was at the descretion of the
photographer. Cars in accessable areas or cars which were of
interest to the photographer may have been more likely to have been
photographed so the list may not be a representative sample.

The UP and Southern conductor's books list every car that the
conductor encountered in their trains and should be more likely to be
representative samples for the territories covered.

John King



--- In STMFC@yahoogroups.com, Malcolm Laughlin <mlaughlinnyc@...>
wrote:

I believe that the random variables in this situation are the
boxcar
counts for each railroad.
================

The box car ownership counts themselves are highly biased. A
significant percentage of the box cars was our of service. For
example, obsolete cars used only in the harvest season for grain
loading to a very limited number of destination cities, cars aiting
for rebuilding, cars designated for shop programs for special
equipment, etc., etc. Depending on the railroad those could account
for anywhere from five to 20 percent of the box cars.

There are other biases in the live cars. Like 50' DD or WD cars
for lumber loading, cars with six foot doors useful for grain but
likely to be rejected by many shippers. There were os many XM box
cars with dimensional characteristics that would keep them from being
randomly distributed.


Malcolm Laughlin, Editor 617-489-4383
New England Rail Shipper Directories
19 Holden Road, Belmont, MA 02478



Anthony Thompson <thompson@...>
 

John King wrote:
I don't think we can draw too many conclusions on car distrubution based on the list of cars in the Charles Collection other than to marvel at the variety of cars which stopped in Harrisburg. The selection of cars to photograph was at the descretion of the photographer . . .
Sure. But you can test any data set, and this one conforms to the Gilbert hypothesis about distribution. Evidently the photographer wasn't terribly selective in his choices. Or, of course, you can believe it's just coincidence.
Conductor's time books are also subject to critique as a data source since not all trains are "average." But again, the tests of those data sets tend to confirm the hypothesis.
You guys who continue to object are going to have to either come up with a basis for something better (other than "I don't like it"), or get used to it.

Tony Thompson Editor, Signature Press, Berkeley, CA
2906 Forest Ave., Berkeley, CA 94705 www.signaturepress.com
(510) 540-6538; fax, (510) 540-1937; e-mail, thompson@signaturepress.com
Publishers of books on railroad history