Thesis (MBA)--Universiteit van Stellenbosch, 1995. / ENGLISH ABSTRACT: Traditional financial ratios are used by different analysts to predict and
evaluate various elements of a company. However, such ratios all suffer from
the basic limitation of accrual-based accounting, hence the perception that
cash flow ratios may be more suitable and possibly even more appropriate to
evaluate companies. Most of these applications employ parametric statistical
procedures of which the validity partly depends on the underlying
distributional properties of the ratio involved.
This study project is an evaluation of the underlying distributional
properties of forty seven cash flow ratios based on the published annual
results of 350 industrial companies, listed on the Johannesburg Stock Exchange
for the period 1975 to 1993. As an initial test, the results of the ratios
were subject to the null-hypothesis of normality as reflected by the p-values
of the chi - square and Kolmogorov-Smirnov tests respectively . Secondly, the
study project attempted to give an indication of the underlying distributional
properties by comparing the average with the median of the various cash flow
ratios.
The first evaluation was done for each ratio. after the data was aggregated,
for both the various divisions and over all the financial periods to evaluate
what is referred to as the sector as a whole. The results of this evaluation
clearly indicate that the underlying distributional properties are definitely
not normally distributed. However, the lack of normality may be founded in
either the lack of homogeneity between the various divisions. or due to the
aggregation of the various financial periods .
A second evaluation was done on the individual ratios for each financial
period. Although the different years lead to slightly different results it
is still apparent that the null-hypothesis of normality for the majority of
ratios are being rejected, especially consistent results to this effect were
found for the period 1988 to 1993.
A third and final evaluation was done on all the different ratios for the
individual divisions. Again the results for the different divisions vary,
but once again it ;s apparent that the null-hypothesis of normality is being
rejected for the vast majority of the ratios . In this regard the results of
specifically divisions industrial holding (15), food (25), engineering (28)
and electronics (29) are particularly consistent.
The lack of normality in the sector analysis may indicate the presence of
specific division relevant characteristics, while the results of the intertemporal
evaluation indicate that the distributional properties approximate
normality by aggregation over time . The analysis also proves beyond any doubt
that irrespective of whether disaggregation is done over either financial
periods or between the different divisions, ratios 02, 03, 06, 15, 16 and 17
are positive skewely distributed . Accordingly, ratios 13 and 27 are negative
skewely distributed .
If one has to advise a researcher or practitioner to whom an underlying normal
distribution of a cash flow ratio is important, one would probably advise in
the first instance to disaggregate over time, as a second app roach to
disaggregate over divisions, but never to treat the industrial sector as a
pooled sample. / AFRIKAANSE OPSOMMING: Tradisionele finansiele verhoudings word deur verskeie markontleders vir
verskillende voorspellings aangewend . Aangesien die verhoudings almal
onderworpe is aan die beperking van opgelope rekeningkundige waarde word
kontantvloeiverhoudings voorgehou as 'n nuttiger en moontlik selfs
doeltreffender metode om ondernemings te evalueer. In die meeste van die
toepassings word parametriese statistiese prosedures gebruik, waarvan die
geldigheid gedeeltelik berus op die onderliggende verdelingseienskappe van die
verhoudings.
Die werkstuk is 'n studie van die onderliggende verde' ingseienskappe van
sewe-en-veertig kontantvloeiverhoudings S005 jaarliks vir die tydperke 1975
tot 1993 gepubliseer is in die finansie1e verslae van 350 maatskappye wat
genoteer is in die nywerheidsektor van die Johannesburgse Effektebeurs. In
die verband is die resultate van die verhoudings eerstens onderwerp aan die
nulhipotese van normaliteit deur dit te evalueer aan die hand van die p-waarde
van sowel die chi-kwadraat- as Kolmogorov-Smirnovtoetse. Vervolgens dui die
werkstuk die onderliggende verdeling aan, deur die gemiddeld en mediaan van
die kontantvloeiverhoudings te vergelyk.
Vir die eerste evaluasie is alle verhoudings oar jie verskillende finansiele
tydperke en van die verskillende afdelings saamgevoeg am die sektor in die
geheel te eva1ueer. Vol gens die evaluasie is die onderliggende verde1ing van
die verhoudings nie normaal nie. Oit mag moontlik logies verklaar word deur
die heterogeniteit van die verskil1ende afde1ings, of weens die samevoeging
van die resultate oar verskillende finansiele tydperke.
Tweedens is die onderskeie finans;ele tydperke afsonder1ik ont1eed . A1 het
die resultate van die onderskeie termyne effens verskil, is dit duidelik dat
die nulhipotese van normaliteit by die oorgrotte meerderheid van verhoud ings
beslis verwerp moet word. Dit is interessant dat in die resu1tate van
spesifiek 1988 tot 1993 die hipotese absoluut konsekwent verwerp word.
laastens is die verhoudings afsonderlik vir elk van die onderskeie afdelings
ontleed. Weereens het die resultate van die onderskeie afdelings verskil,
maar dit was oak weer duidelik dat die oor9rote meerderheid van verhoudings
die nulhipotese van normaliteit verwerp . In die verband was veral die
resultate van die afdelings nywerheids beherend (15), voedsel (25).
ingenieurswese (28) en elektronika (29) merkwaardig konsekwent.
Die gebrek aan normaliteit tussen die verskillende afdelings mag maontlik
aantoon dat sekere eienskappe, eie aan die afdeling, die verdelingseienskappe
beinvloed . Die resultate van die intertemporale studie toon beslis dat
normaliteit benaderd word indien verhoudings oar tyd saamgevaeg word. Wat die
ontleding ook bo alle twyfel bewys is dat die verhoudings 02, 03, 06, 15, 16
en 17 vir die sektor as 'n geheel, die intertemporale- en die
kruissektorale-ontledings konsekwent onderliggend positief skeef verdee1 is.
Verhoudings 13 en 27 is egter onderliggend negatief skeef verdeel.
Aanbevelings rakende die onderliggende verdeling van kontantvloeiverhoudings
is beslis om eerstens verskillende finansiele tydperke te onderskei en
tweedens die verskillende afdelings. maar beslis om nie die nywerheidsektor
as 'n geheel te evalueer nie.
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:sun/oai:scholar.sun.ac.za:10019.1/54818 |
| Date | 03 1900 |
| Creators | Groenewald, Jakobus Johannes |
| Contributors | Hamman, W. D., Stellenbosch University. Faculty of Economic and Management Sciences. Graduate School of Business. |
| Publisher | Stellenbosch : Stellenbosch University |
| Source Sets | South African National ETD Portal |
| Language | af_ZA |
| Detected Language | Unknown |
| Type | Thesis |
| Format | 117 p. |
| Rights | Stellenbosch University |
Page generated in 0.0016 seconds