8/11/2019 Forecasting Security Prices 1925
1/7
Forecasting Security Prices
Source: Journal of the American Statistical Association, Vol. 20, No. 150 (Jun., 1925), pp. 244-249Published by: American Statistical AssociationStable URL: http://www.jstor.org/stable/2277120.
Accessed: 26/04/2014 20:28
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at.http://www.jstor.org/page/info/about/policies/terms.jsp
.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].
.
American Statistical Associationis collaborating with JSTOR to digitize, preserve and extend access toJournal
of the American Statistical Association.
http://www.jstor.org
This content downloaded from 69.123.206.101 on Sat, 26 Apr 2014 20:28:18 PMAll use subject to JSTOR Terms and Conditions
http://www.jstor.org/action/showPublisher?publisherCode=astatahttp://www.jstor.org/stable/2277120?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/2277120?origin=JSTOR-pdfhttp://www.jstor.org/action/showPublisher?publisherCode=astata8/11/2019 Forecasting Security Prices 1925
2/7
244 American tatisticalAssociation
[82
It is this xtensionrmodificationf hePearsonian ethod hichMr.Peabody
used
n
fittingoth
he
Gompertz
nd
theMakeham urves.
The discoveryhat he use of an extramoment ill liminatehetranscen-
dental ermsthe utcomef search n thepart fMr.Sasuly or method f
fittinghe xponential be- mwithoutaving oreducet o ts ogarithmicform
and
without aving o solve transcendentalquation
n
m.
When hepresent
writeralledhis
ttention
o the
fact
hat henegativexponentials oneof he
uncommonypes fPearson's ystem ffrequencyurves nd that n fitting
this urve
y
themethodf
momentsearson asrecourse eithero ogarithmic
transformationor othe olutionf transcendentalquation,' r.Sasuly, fter
an examinationf
Pearson'smethod,ame o the onclusionhat ny ranscen-
dental erm
may
be
eliminatedhroughhe use of an extramoment.At
the
time hatMr.Peabody's rticlewas being repared,he cope nd validityf
this
xtensionf hemethodfmoments ere
nly artiallynvestigated
hichs
one
of
he
reasonswhy hepublicationfthismethod as deferred.
Mr.Peabody indshat he ess nvolvedummation ethod ivesmore atis-
factoryesults han his modified) ethod f moments.This s due either o
the
seof aw
momentsr othe
omparativelyarge robablerrorf he extra)
fourth
oment.
Should his
inding
e
confirmedy wider xperienceith
he
two
methods,
e
will,
f
ourse,
ave o
conclude
hat he
dvantage
f
using
n
extramoment
s of
more heoreticalhan
practicalmportance.For thetime
beingt s best osuspendudgment.
HENRY SCHULTZ
Washington,
. C.
FORECASTING SECURITY PRICES
On
April
17,
1925,
dinner
meeting
f
he
American
tatisticalAssociationwas
held
at
the
Aldine
Club
in
New York
City. About
420
people were
n
attend-
ance. The topicunderdiscussionwas Forecastitigecurity rices.
The first our
speakers represented ommercial
gencies active
in
the
fore-
castingfield.
Mr.
Paul
Clay
of
Moody's Investors' ervice scribed
he success
of that nstitution o its creed.
He affirmed
belief
n
economists
nd
statisti-
cians
because they
are far
ahead of
the
average businessman
in
realizing
he
practical value of economic
tudy,
and
they have done wonders n
learning o
understand he
ways
of trade
and
in
reducing
ts
hazards.
He
continued:
The free layof
economie orces
as
never etterafeguarded.
This
country as
nevermorefree
from
ctual
monopolies.
The stock
market
was never
moreun-
spotted y
manipulation,
nd bulls
and
bears
neverbefore ad
so
little o do with
your wn nvestmentuccess s theyhave now.
We
have never een he
nfallibility
f
udgment
hich s
supposed
o
belong o the
mythicalnsiders. To
our
mind,
hebelief
n
nside
nterestss
as
fabulous nd anti-
quated s that n theoracle f
Delphi.
1
The
exponential
may
be written
=
whereN
is the total
frequency
nd
cr
s
the
stand-
ard deviation.
See Elderton,W. P.:
Frequency
urves
nd
Correlation,
ddendum, 1917,p.
8,
and
Phil. Trans.,A,
vol.
ccxvi,pp. 429-457.
This content downloaded from 69.123.206.101 on Sat, 26 Apr 2014 20:28:18 PMAll use subject to JSTOR Terms and Conditions
http://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsp8/11/2019 Forecasting Security Prices 1925
3/7
831
Notes 245
We believe
n
the
rule
of
economic
orces,
nd
this
s whywe reject
he
Delphic
oracles nd inside nterests.There s nothingmysterious
bout the stockmarket
except hosephases f
twhich esearchnd datahavenotyet overed. It is merely
the greatAmerican apital
market,ntowhich lows he
surplus apitalof every n-
dustry, nd out of which low treams f capital to supply
he expansionneedsof
every ndustry. Typicalmen,whateverheir usiness nd whereverhey ive,buy
stocks nd bondswhen
heyhave surplus unds,
nd
not
t other imes. Theydump
securities n the markets
when hey re
in
financial
istress, nd theyneverdo so
when hey re
not n distress.
Therefore,
hefinancialituation
f
the typical usi-
nessmandominateshecourse fthemarket.
So it sthat he tockmarkets
a
creature f very ay
economic
orces.
Complete
data wouldprobably how
hatduring ullmovementsherise s roughly roportion-
ate to the nflow f apital avedout of he arnings f
ndustries. t wouldprobably
show
hat
during
earmovementshe decline
s
proportionate
o theurgentiquida-
tionon thepartofbusiness
menwho are short fnecessary
apital. Presumably,f
we
had an accuratemeasure
f
the surplus arnings
f
ndustries
n
prosperous
imes
and
of hecapital hortages
f ndustries
n
depressionimes,
we should now xactly
whatthe tockmarkets going o do.
Ourresearches ave
disclosed
many vidences
f
herule f conomic
orces. Per-
haps themost triking
s theparallelismfour radebarometer
ith hemain wings
of the stock market.
This
barometers
a
monthly
eighted verage
of
selected
barometricalrade
eturns;
nd we havecompiled
t n
either
inishedr
n
crude
orm
*back
o 1863.
Throughout
he entire ourse
t runs
generally
arallel
o the main
swings
fthe stockmarket.
Close studyreveals hefact that the stockmarket
s a
creature
f
trade,
nd
a
trailer fter t, and not
a
leader
or
a
maker
f trade
onditions.
The popular elief
thatthestockmarket iscountsr anticipates as no basis
n fact. The belief tself
doubtless
risesfrom he fact thattradeconditions re so complex
hat
very
few
peoplerealize hanges
herein
ntil
months
fter
hese hanges
ave occurred.
The high eaksof he tockmarket re theresult f nunbrokenhain f conomicforces. First, he money alue of our commercialurnoverominates he trend f
interest ates; econd, ommercialaperdiscountsmove
with nterestates, nd bond
yields ollow ommercialaper. The highpeak
n
the
tockmarkets thatpoint
t
which
rices
ave risen o far
hat
tock
yields
re
slightly
elowbond
yields.
This
has been consistently
rue
of
the
majorpeaks
n
the stock
market
hroughouthe
history
fthe New
York
StockExchange.
Seasonal variations
f trade and stock
prices
re
practically
like except
n that
portion
f
the year during
which
he money
market
xercises
more nfluence pon
securities
han
he
rade ituation oes. From hemiddle
f
eptember
o themiddle
of
January,
he
stockmarket esponds
o the
money
market
urve,
whereas
uring
theentire alance
of
he
year
t
responds o thegrowthf
rade
onditions.
Further-
more,
he
money
market urve tself s the creature fthe trade urve.
Now, fthe stockmarkets a creature feconomic orces,t followshattheright
method f
forecasting
s
to
study hese
orces.
By way
of
defining
he
rightmethod,
let
me first
peak
of
ome
f hemethods hichwehaverejected. First, here
s
the
single
ndexmethod.
Many years go
it
was
believed
hatpig
ron
was
a
good
fore-
caster
or
he stockmarket,
nd now
manypersons
hink hatone can forecast uc-
cessfullyustthrough
bservationfblastfurnace apacity,
ndnothing lse. There
is
no question hat hese
ndices reworth tudying,utwe do notbelieve
n
any ingle
index,
or
he ourse f he
tock nd bondmarketss a reflectionfthe change
n
the
financialonditionftheentireAmerican eople. If the
resultant f forces an be
calculated
rom
single actor
n
finance, hynot
n
physics,
s well?
Anotherorecastingnstrument
hich
we haverejected
s
the utomatic
arometer.
There re those hat, ycombiningertain arometrical
eturns,btain barometer
or ndexnumber hich s supposed o move headof he tockmarketnd ndicate tscourse. Thissurelys better han he ingle ndex, uthere gainthere s the erious
objection
hat
theforecaster
s
compelled o reject
ll
economicorces ot
ontained
in that
barometer.
Doubtless,however,
he
most
popular
method f
forecasting
s
chart
eading,
nd
this, oo,
we
have rejected.
The chart s
a
post-mortem
nd
not
a
diagnosis,
t is a
result,
ather han
a
cause,
and
its occurrence
s
afterward
ather
han
beforehand.
Prices, ccording
o
John
tuartMill,are founded pon upply nd demand, nd not
upon
double
ops
nd
double ottoms. Adam
Smith
elieved
hat he
auses
of
value
This content downloaded from 69.123.206.101 on Sat, 26 Apr 2014 20:28:18 PMAll use subject to JSTOR Terms and Conditions
http://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsp8/11/2019 Forecasting Security Prices 1925
4/7
246 American
tatisticalAssociation
[84
wereutility nd labor, nd
not
bull and bear chart urves. We
have chosen o cast
our ot and ourfortunes ith hese ober tudents f economic orces.
Now, since he course f
the stockmarkets the resultant feconomic orces,et
me tell you what omeof these
measures re
and
howwe use them.
First, here s
our trade barometer, hich s a weighted verage of barometricalrade returns.Trade conditionsre bullish rbearishnaccordance ith hemovementf his rade
barometer,xcept hat n excessive
ise ar bovenormals bearish, hile n excessive
decline arbelownormal s bullish.
Anothermeasure f conomicorces hichwe highly alue s our
mploymentndex.
Its value lies in the factthat
t is a crudemeasure f the purchasing
ower f the
American eople. When abor
s well employed, lantcapacity
nd capital s also
wellemployed,nd employerss well s theirmen re making
money.
A third
mportant easure
feconomic orces mployed y us is theratioofbank
loansto deposits. A sharp ise n these atios nvariablymeans hat
apital upplies
are
running hort, nd vice versa.
Any doubt as to interpretationan readily e
eliminated y use ofourmonthlyverages f nterest ates.
Another f our measures
feconomic orcess our nflationndex,
which s an ab-
stractndex umbero constructedhat twill ise harply heneverredit,nventories
or
operating osts re nflated.
By inflation e mean
not
mere
xpansion, owever
great, ut any condition funsoundnessr overexpansion hich s
bound o result n
forced iquidation n thepart
ofbusinessmen.
Of
ourothermeasures f conomicorces,
willmention
nly
ne. This
s an
index
number f the atent arning
ower f ndustries. By latent arning
ower, mean
theearnings hichmight heoretically
e made fmaterials,abor,
apital,
tc., ould
be bought t current rices, nstantly
onvertedntofinished oods, nd the goods
instantly old. In our woolen
ndex, or xample,we figureust
as a woolen
mill
managermight. We compute
he cost ofproduction,nd compare
with
hemarket
value of thefinishedoods.
A
few pplications fthese
measures feconomicorceswill erve
o
illustrate
ur
methods fforecasting.At thepresent ime, or xample, e arenot aking bearish
position. On the 5th and 12th f March we advisedprofit
aking,merely ecause
stocks
ad risen oo fast
s compared
ith rade
mprovement,
nd themarket
as
n
an
overboughtondition. However,
hesemeasures fforces avenot urned earish,
and
we haveregarded
he
decline
which ccurredince
he
middle
fMarch s
nothing
worse
han
fair
pportunity
o
repurchase
tocks
t
substantial
oncessions.
Mr.
Clay
further
ointed
out that
in
1920 forecasts f his servicewere
based
primarily
n the
overexpansion
f
credit.
In
the summer
f
1921, they
were
dominated
by
the latent earningpower
of industries.
In
October, 1922,
the
employment
ndex was the
governing actor,
while
n
December,
1924,
reliance
was placed mainly upon the employmentndexsupplemented y the ratio of
loans
to
deposits.
Mr.
Clay
concluded
his
remarks
with
plea
for
more xtensive esearch
n
this
field.
The
next
peaker
f
the eveningwas
Mr. WilliamPeter
Hamilton,
ditor fthe
Wall
treet ournal.He set forth he
theory
f the stock
market, riginal
with
the
ate
Charles
H.
Dow,
former ditorof the
Wall
treet
ournal.He
said:
Themajor wing
f
thestockmarket
s
themost
mportant
fthethree endencies
intowhich ow's
theory
ivides he
price
movement.
Over
period
f
twenty-one
years, tarting
rom
une,
900,
nd
ending
with
June, 921,
herewere welve
major
movementsf he tockmarket-sixupandsixdown. What sinterestingo note s
that
the upwardmovement,
r
bull
swing, veraged wenty-five
onths
n
length,
while
he
downward,
r bear
market, veraged
eventeen
months.
It
is to be noted
hat
hese ull
and bear
markets
re
neveruninterrupted.
ere
comes
n
what
may
be
called
the
secondary
movement f the Dow
theory.
A
bull
market
s
nterruptedy sharp
eactions,s,
for
nstance,
hat
which
et
n on
the7th
of ast
March, eaching
ts ow
point
n
March
30,
while bearmarket
s
checked
n
ts
progress y sharp ecoveries,
fonly
or
he echnical
eason
hat
uch
markets
apt
to become versold.
This content downloaded from 69.123.206.101 on Sat, 26 Apr 2014 20:28:18 PMAll use subject to JSTOR Terms and Conditions
http://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsp8/11/2019 Forecasting Security Prices 1925
5/7
85]
Notes
247
That s the Dow
theory
ftheprice
movement.
t
does
not
claim
o
be
anything
more han an
hypothesis,nd
practicallyvery cientific
iscovery
he
world ver
saw
developed rom
workable
ypothesis.
The speakerexpressed heview that thepresent eaction fthe stockmarket
was
merely minor
hase
of
general
pward
movement
which
was
not
yet
over.
Mr.
Hamilton
pointed out that
he
had
published
a
book
in
1922,
called The
Stock
Market
Barometer,
howing
he
application
of the
principles
f the Dow
theory. He
stated
that
if
the
Harvard
UniversityCommitteeon
Economic
Research
were to
omit
bank
debits
from
ts
line
of
speculation, hat this
line
would
correspondo
the
Dow-Jones
verages used by
the
Wall
StreetJournal.
He
stated
that the line of
speculation
lwaysprecedes
by
some
months
cor-
responding
luctuation
n
the
line
of
business. Mr.
Hamilton
emphasized
he
factthatmanipulation f themarkethad little ignificance. He stated, It is
not
difficulto see
how,
n
a
narrowmarket
with a
small
volume of
trading,he
daily fluctuation
ould be
manipulated. It
would
be
much more
difficult
o
manipulate
the
secondary
reaction or
rally,
because
it would be
necessary o
take
n
hand
at least
a
majority f
forty
ctivestocks.
You
maytake t
from
me
that
there s
no financial
nterest, r
combination
f
financial
nterests,
n
Wall
Streetor n
the
whole
country,
hich
s bigenough
o
do anything f
the
kindor
foolish
nough
o try.
Mr.
Hamilton
cited
as
an
exampleof the
surprising ay in
which the stock
marketforecasted usinessthat in December,1917,elevenmonthsbefore he
armistice,
he
market
redicted he final
nd
complete efeat
f
Germany.
Mr.
Hamilton
vigorously
ondemned
stock
tipping
gencies.
His
remarkswere
followed ythose
of
RogerW.
Babson, of
Babson's Statisti-
cal
Organization,who
llustrated
is
remarks y
the aid
of a chart,
howing
he
equality
of
action and
reaction
n
business
for long
period
of years.
He
held
that
while t
was
folly o
attempt
o
forecasthe
short
wings
n
the
stock
market,
practically
ll
the
economic
ervices
have a
cleanrecord
n the
ong-swing
move-
ments. He
said that we
know
whatperiod
we
are in and
what the
next
period
willbe; the onlyuncertain actorbeingthe durationof the respective eriods.
As the
Department
f
Commerce nd
other
rganizationsre
continually
ollect-
ing
and
compiling
moreand
more
statistics, ll
the
services re
constantly
m-
proving
heir
bility o
determine he
duration fthe
periods. He
pointedout
that
the
science
of
forecasting as
rapidly
developing
nto
a
very ound
and
use-
ful
industry,
nd
that
everyadditional
person
who
becomes nterested
n this
study
makes the
next boom
less
reckless
nd the
nextpanic less
severe.
Mr.
Babson
predicted
hat
the
forecastingervice
f
yearshence
would
combine
he
good features
f
the
eading
gencies
hat
now exist. He
also
expressed heview
thatwe are still n theupwardmovement fthe stockmarket.
The
fourthpeaker
was
Mr.Ray
Vance,
President f
the
Brookmire
conomic
Service.
He
began
bydefininghe
term,
forecasting
s
being sucha
measure
ofbusiness
happenings
nd
such
a
survey
f
theresults
whichhave
followed hese
happenings
n
the
past,
s
will
permit
he
nticipation f
ecurity rice
movements
with
a
worth-while
pproximation
o
accuracy.
He
emphasized
hefact
that
by
security rice
changes
he
did
not mean
the
changeswhich
ccurred rom
our
This content downloaded from 69.123.206.101 on Sat, 26 Apr 2014 20:28:18 PMAll use subject to JSTOR Terms and Conditions
http://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsp8/11/2019 Forecasting Security Prices 1925
6/7
248 American tatisticalAssociation [86
to hour, r even frommonth o month, ut those major movementswhich ordi-
narily ccur not oftener han once a year. He also stressed he point that the
Brookmiremethodcannot be regarded s
in
its finalform. Changes of major
importancemay be made at any time. He stated that the factors sed and the
reasonsforusing
them re:
1. The height f stockprices hemselvesnd the activity n the exchange. The
higher tocks re,themore ikely other hings eing qual) they re to decline, nd
incidentally,he greater heactivity, he more ikely s a decline.
2. The greater he volume f basic raw material roduction oing n within his
country,
he
greaters the chance
hat
productions runninghead of consumption.
Therefore,igh physical olume f productions used as a bearish actor, nd low
physical olume s
a
bullish ne.
3. The mportationfgoods s, as an economic act, n equivalent
o
an increase
n
productionnd viceversa. Therefore,heratiobetweenmportsnd exportss used
as a third actor. When his atio s high, t indicates rouble, nd when t is low, tshows fundamentallytrong osition.
4.
The
turnoverfbankdepositss measured y theratio fbank learingso bank
deposits.
A
high
ate of
turnovers unfavorablend
a
low
rate
favorable.
5.
The
supply
f
purchasing ower
n
theform fbank deposits s
a
flexiblehing
until hebanking trengthf he ountrys exhausted, runtil governingody, uch
as ourFederalReserve oard, rtificiallyightensredit. As an ndex fwhetherhis
pointhas beenreached r not,we use thecommercialaper ate
n
theUnited tates.
6.
Because the
supply f oanablebank funds s really n internationalffair, e
used
as
a
sixth actor he
open
market
money ate
n
London.
Mr. Vance
pointed
ut
that
while
his
method
has
fallen
onsiderably
hort
f
100 percentaccuracy, hat t has been very uccessful uring he past two years.
Recent
changes
n
the method
re
expected o
make it
even
better
han
it
has
been.
He
submitted
chart n which
were
marked
he
points
t
which lients f
the
service
had been advised to
buy
and sell
stocks,
nd he
emphasized
he
fact
that
all
predictions
made
by
the
Brookmire
ervice
were
pecific. He stated
that
in
February advice was given to sell stocks, and that therewas no reason to
believe that this advice
was not sound. He
concluded
his remarks
y saying,
We consider he
service,
n
spite
of ts
lack
of
presentperfection,
as accom-
plished
ar
n
excess
of what is accomplished y the average buyer of securities,
and hence ts use is justified.
The
program
was closed
by
Mr. Frederick
R.
Macaulay
of
the National
Bureau
of Economic
Research,
who
gave
a critical
eviewof the remarks f the
preceding peakers. He began by pointing ut that the Dow theoryof fore-
casting ndorsed y
Mr. Hamilton
tood
n
contrast o the other hree heories
n
that
t was based
upon
the
movements
f the
stockmarket tself. He
observed
that
therewas
a
striking imilarity
etween
he
fluctuations f the
stock
market
and those
of chance
curvewhich
may
be obtained
by throwing
ice.
Everyone
will
dmit hat he
course
f such a
purely
hance
curve
annotbe
predicted.
If
the tockmarket anbe forecast rom graph f tsmovements,tmustbe because
of
ts
difference
rom he
chance
curve.
There
undoubtedly
re
such
differences,
said
Mr.
Macaulay.
Because of
the
earning ower
of
the
companies hemselves,
for
example,
t
is
certain hat
the
curve
will
never
reach the
zero line
and
will'
never
rise
ndefinitely.
There is
no
limitto the
distancewhicha chance curve
may
rise
or
fall.
Mr.
Macaulay expressed
he
view, however,
hat
Mr.
Hamil-
ton's
method of
prediction
was
much
less
dependable
than
those
used
by
the
This content downloaded from 69.123.206.101 on Sat, 26 Apr 2014 20:28:18 PMAll use subject to JSTOR Terms and Conditions
http://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsp8/11/2019 Forecasting Security Prices 1925
7/7
87]
Notes
249
other
hree
peakers,
nasmuch
s the ssence
f heDow
method
asto buy
s
long s the
marketontinuedo rise nd ell s
long s it
continuedo fall,which
was ust
thereverse
f
taking
dvantage
f
hefact hat he
market
as a
defi-
nitebottom ndthatthere s a limit o thedistance o whicht canrise. Mr.
Macaulay
lso
uggestedhat t
might e
unwise,
n
ttempting
oforecastrom
record
f he
market
tself,o
exclude
hevolumef rading rom
onsideration.
He called ttentiono the
fact hat
'double
ops
and 'double
ottoms seem
to be ust as characteristic
f
chance urve s ofthe tock
market ecord,nd
hence hat t is doubtful
hetherheyhave
nyparticularignifications
pre-
dicting actors.
He next
urned is
ttentiono
the
methods
sedby the
ther hree
gencies;
namely, hat of
utilizing
ecordedmovementsf economic
orces utside
he
marketso howwhat s goingohappen o tock rices. Heraised hequestion,
in
regard o Mr.Babson's
hart, s to whether
t was not ustas difficulto pre-
dict he reas
bove nd below hetrend, s
to forecasthe
course fthe
tock
market irectly.
Mr.
Macaulay howed hat he hief ay n
which he
method
used
by
Mr.
Clay differed
rom hatusedby Mr. Vance
was that
Mr.
Vance
attempts
o
constructmathematical
ormula
orweights, hile
Mr. Clay
uses
mental
ather
han
ny
mathematical
eighting.Mr.
Clay, moreover,akes
into
onsiderationny vent r
force hich edeems
mportant,hileMr.Vance
tends o restrict
imselfo the ix
factors
n
hisbarometer.
The apparent
d-
vantage futilizinglltypes fforcessperhapsmore han ffsetythedanger
that
he
forecaster
ill e
undulynfluencedy events f he
moment,
nd
may
thereforese
weightshat renot ogical.
The
meeting
as ablypresidedver yMr.
Walter . Case,Presidentf
Case,
Pomeroy
Co.,who tated hat he xperience
fhisfirm ad
favoredhe
useof
outside
conomicactors
n
foretellinghe
general ourse fthemarket. As to
thepresent
utlook or he
prices f ailway
tocks, ewas nclinedo agree
with
the
first
hree
peakers, ho
believed hat he isingmarket ould
ontinue,
ven
though
e
had
great espect or he method sed
by
Mr. Vance
whichgave
contraryndications.
THE
COST OF
LIVING INDICES OF
THE N. I. C. B.
AND OF THE
B. OF L.
S.
In her
xtendedrticlen
theDecember umber f heJOURNAL,
Miss Elma
Carr
of the Bureau
of Labor
Statistics
egins
er
valuable
description
f
the
differences
etween hecost
of iving ndices
f
the
Bureau nd the
Boardwith
a referenceotheir xtensive se nwage djustmentsnd endswith hecon-
clusion hat
hemethods f
theBureau re far
superior nd that
ts ndex
should
hereforee
consideredmuchmore eliable.
If
the
JOURNAL'S
pace
permitted,
nd
f
he
ubjectwere mportantnough ot
o havebeen
forgotten
sinceMiss
Carr's rticle ppeared,
greatdeal could
be said about t. It is
possible,
owever,o venture nly few
general
omments.
This content downloaded from 69.123.206.101 on Sat, 26 Apr 2014 20:28:18 PM
http://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsp