International Journal of

Financial Studies

Article
Exploring the Dynamics of Profitability–Liquidity Relations in
Crisis, Pre-Crisis and Post-Crisis
Piotr Ratajczak * , Dawid Szutowski and Jarosław Nowicki

Department of Controlling, Financial Analysis and Valuation, Poznań University of Economics and Business,
al. Niepodległości 10, 61-875 Poznań, Poland; [email protected] (D.S.);
[email protected] (J.N.)
* Correspondence: [email protected]

Abstract: The aim of this study is to verify the stability of the profitability–liquidity relationship over
time, as well as to determine this relationship in terms of its level and structure. In this context, three
main research questions were formulated. First, is the profitability–liquidity relationship stable in
times of crisis? Second, what is the profitability of companies with high and low liquidity? Third,
what is the liquidity of companies with high and low profitability? This study uses a self-organizing
map (SOM), a data visualization technique that is a type of artificial neural network trained in an
unsupervised manner. A dataset covering the period from 2019 to 2021, consisting of 300 Polish
companies from the wholesale and retail sectors, was used. The main results of this study indicate
that: (1) companies with a balanced profitability–liquidity relationship in the pre-crisis period (2019)
maintained this relationship in the crisis (2020) and post-crisis periods (2021); (2) companies in the
clusters with the relatively highest and lowest profitability have the relatively lowest and moderate
liquidity both before and after the crisis period; (3) the majority of companies during non-crisis
periods demonstrate that profitability is not reliant on liquidity, suggesting an absence of a clear
relationship; (4) in the post-crisis period, companies with the relatively lowest operating cash flow
margin (OCFM) exhibited the relatively highest net profit margin (NPM) and other profitability
ratios, as opposed to the pre-crisis and crisis periods. This study fills the gap resulting from the
incomplete—most of all static—understanding of the relationship between profitability and liquidity.
Citation: Ratajczak, Piotr, Dawid Moreover, this study employs a self-organizing map (SOM) which has not been used in the literature
Szutowski, and Jarosław Nowicki. regarding the research area undertaken.
2024. Exploring the Dynamics of
Profitability–Liquidity Relations in Keywords: profitability; liquidity; crisis; self-organizing map; SOM; Kohonen’s map; artificial neural
Crisis, Pre-Crisis and Post-Crisis.
network; ANN
International Journal of Financial Studies
12: 16. https://doi.org/10.3390/
ijfs12010016

Academic Editor: Zied Ftiti 1. Introduction

Received: 8 December 2023
Explaining the origins of performance disparities between companies is a fundamental
Revised: 31 January 2024
theoretical and empirical concern, particularly within the domains of corporate finance and
Accepted: 1 February 2024 strategic management. “Although different theories have tried to illuminate the reasons
Published: 10 February 2024 why some firms are more profitable than others, and a large amount of research has
considered and explored different factors that may impact firm performance, the issue of
firm profitability continues to be an actual, significant and inexhaustible phenomenon that
attracts the attention of many researchers and practitioners” (Pervan et al. 2019, p. 968).
Copyright: © 2024 by the authors. The literature suggests that a firm’s internal factors have a greater influence on its perfor-
Licensee MDPI, Basel, Switzerland. mance than external ones (Hawawini et al. 2003; Makhija 2003). The debate typically focuses
This article is an open access article on liquidity as a crucial factor that affects short-term performance. However, empirical
distributed under the terms and studies on the relationship between liquidity and short-term performance produce contra-
conditions of the Creative Commons
dictory outcomes. Some studies have confirmed a positive relationship (Hirsch et al. 2014;
Attribution (CC BY) license (https://
Nunes et al. 2012), no relationship (Jaworski et al. 2018; Sur and Chakraborty 2011) or—most
creativecommons.org/licenses/by/
often—a negative relationship (Eljelly 2004; Maçãs Nunes et al. 2009).
4.0/).

Int. J. Financial Stud. 2024, 12, 16. https://doi.org/10.3390/ijfs12010016 https://www.mdpi.com/journal/ijfs

Int. J. Financial Stud. 2024, 12, 16 2 of 18

Holding liquidity in the short term may reduce profitability, but it could ultimately
improve it in the medium to long term. The bolstering effect of increased liquidity is
rooted in healthy fundamentals that enhance stakeholder confidence and reduce the risk
of bankruptcy (Nanda and Panda 2018). Interestingly, Vieira (2010) ascertained that over
the medium term, the majority of companies sampled were unable to maintain both
high liquidity and low profitability. Conversely, those companies with high liquidity and
profitability—or low liquidity and profitability—tended to remain in these states.
The relationship between liquidity and profitability remains incompletely understood,
the reason for which may be the factors that disrupt this relationship. One factor to be
explored is the concept of crisis. While prior studies have looked into how companies per-
formed during crises, none of these studies have delved into the specific variations in com-
pany performance comparing pre-crisis, crisis, and post-crisis periods (Cheong et al. 2021).
This study fills the gap resulting from the indicated incomplete understanding of
the relationship between profitability and liquidity. Moreover, this study employs a self-
organizing map (SOM)—the data visualization technique being a type of artificial neural
network trained in an unsupervised manner—which has not been used in the literature
regarding the research area undertaken, enabling conclusions to be obtained that explain
the partially disparate research results presented in the literature. A dataset spanning from
2019 to 2021, comprising 300 Polish companies from the wholesale and retail trade sector,
was used.
The aim of the current study is to verify the stability of the profitability–liquidity rela-
tionship over time, as well as to determine this relationship in terms of its level and structure.
This study enhances the academic discussion and has implications for business practice.
The results of this study indicate that companies with a balanced profitability-liquidity
relationship in the pre-crisis period maintain this relationship during the crisis and in the
post-crisis periods, although the levels of profitability and liquidity ratios decline. Secondly,
the results of this study show that companies in the clusters with the relatively highest
and lowest profitability have the relatively lowest and moderate liquidity both before and
after the crisis of 2020. On the other hand, the group of companies with the relatively
highest liquidity tends to have relatively moderate profitability. Moreover, concerning the
majority of companies during non-crisis periods, the relationship between profitability and
liquidity aligns with the midpoint of Gentry’s curve (Gentry 1976). Thirdly, an interesting
phenomenon emerged in the post-crisis period of 2021, with companies having their
relatively lowest levels of operating cash flow margin (OCFM) while also having their
relatively highest levels of net profit margin (NPM) and other profitability ratios, and vice
versa. This phenomenon is particularly striking when compared to the years 2019 and 2020
characterized by a “normal” convergence of OCFM, NPM and other profitability ratios.
The structure of the paper is as follows: Section 2 presents the literature review and
research questions; Section 3 describes the methodology used in this study (including
sample, variables and analytical approach); Section 4 shows the results of the research,
and finally, Section 5 is devoted to discussion and conclusions, as well as the future
research directions.

2. Literature Review and Research Questions

Maximizing profit while ensuring sustainability is crucial to a company’s long-term
survival. The literature has extensively examined the changing trends in corporate prof-
itability over time and scrutinized the influence of internal (firm-specific) and external
(industry-specific) factors on this phenomenon (Nanda and Panda 2018). As far as the-
oretical basis is concerned, the researchers are engaged in a debate regarding whether
a company’s performance is primarily shaped by its internal resources, as posited by
the resource-based theory, or by external factors emanating from the broader economic
environment, as suggested by system theory (Cheong et al. 2021).
Int. J. Financial Stud. 2024, 12, 16 3 of 18

The literature suggests that a firm’s internal factors have a greater influence on its
performance than external ones (Hawawini et al. 2003; Makhija 2003). The debate typically
focuses on liquidity as a crucial factor that affects short-term performance. However, em-
pirical studies on the relationship between liquidity and short-term performance produce
contradictory outcomes. Some studies have confirmed a positive relationship (Hirsch et al.
2014; Nunes et al. 2012), no relationship (Jaworski et al. 2018; Sur and Chakraborty 2011)
or—most often—a negative relationship (Eljelly 2004; Maçãs Nunes et al. 2009), poten-
tially because of the cost savings in maintaining liquid assets and using cheaper funding
sources. However, this negative relationship could potentially manifest only in the short
term. In the medium to long term, inadequate liquidity may lead to reduced profitability,
augmented loan demands, and insufficient cash flow, thereby engendering a cycle that
adversely impacts a company’s performance (Hirigoyen 1985).
Holding liquidity in the short term may reduce profitability, but it could ultimately
improve it in the medium to long term. The bolstering effect of increased liquidity is
rooted in healthy fundamentals that enhance stakeholder confidence and reduce the risk of
bankruptcy (Nanda and Panda 2018). Indeed, many studies suggest that higher liquidity
and profitability decrease the likelihood of firm failure (Ratajczak 2023).
Interestingly, Vieira (2010) ascertained that over the medium term, the majority of
companies sampled were unable to maintain both high liquidity and low profitability.
Conversely, those companies with high liquidity and profitability—or low liquidity and
profitability—tended to remain in these states. Byczkowska and Kuciński (2018) observed
that over the long term, a significant level of reported profits feeds into a company’s cash
flow, and this has a positive impact on their ability to pay. This could potentially explain
the phenomenon of high liquidity and profitability maintained by certain companies.
Kowerski (2016) contends that assuming a linear correlation between profitability and
using static liquidity ratios does not comprehensively reflect the topic at hand. Instead,
a relationship that follows an inverted U-shape may be more appropriate (Gentry 1976).
This implies that profitability is not favored with either very high or very low liquidity
levels. When liquidity values are small, boosting liquidity could enhance profitability
until a certain threshold is surpassed. However, once that threshold is passed, increasing
liquidity may ultimately lead to decreased profitability.
The relationship between liquidity and profitability remains incompletely understood,
the reason for which may be factors that disrupt this relationship. One factor to be explored
is the concept of crises. While prior studies have looked into how companies performed
during the global financial crisis, none of these studies have delved into the specific
variations in company performance comparing pre-crisis, crisis, and post-crisis periods
(Cheong et al. 2021). However, Lee et al. (2017) highlighted that crises affect company
profitability by influencing various internal factors. Vieira’s (2010) empirical research on
a group of publicly listed Portuguese companies indicates that economic hardship has
a detrimental impact on the business outcomes of non-financial companies. Meanwhile,
Adelopo et al. (2018) separated three periods of pre-crisis, crisis and post-crisis in order to
examine bank profitability in the Economic Community of West African States.
The observed diversity in how profitability and liquidity relate suggests the presence
of factors that influence the subject relationship. However, only a limited number of authors
have explored this phenomenon. Jaworski and Czerwonka (2021) discovered two notable
moderators that significantly influence both the strength and direction of the relationship
between profitability and liquidity: private sector lending and the development of capital
markets. Undoubtedly, these factors are potentially linked to the crisis phenomenon.
Consequently, there remains a lack of clarity regarding the impact of both industry-specific
and firm-specific factors on a company’s performance, especially in times of crisis.
A lack of understanding of the stability of the profitability–liquidity relationship leads
to partially incompatible results, as the inclusion of pre-crisis, during the crisis and post-
crisis periods in a research sample distorts the results of such research. This is of particular
practical importance because, as Chow et al. (2018) point out, macroeconomic uncertainty
Int. J. Financial Stud. 2024, 12, 16 4 of 18

such as the global financial crisis poses a significant challenge to companies’ resource
allocation decisions.
The stability of the subject relationships is a crucial consideration, particularly in light
of the current economic climate. Many economic principles are being challenged due
to vastly different macroeconomic conditions. For instance, the Federal Reserve Board’s
policy of injecting liquidity into the financial system lacks a scale of comparison. While
some authors have addressed this issue, they have only done so in relation to banks and
other financial institutions, not businesses. As demonstrated by the recent pandemic
crisis, economic principles remain effective. However, it may take more or less time
to translate one phenomenon into another. Research on the stability of the relationship
between profitability and liquidity will enhance comprehension of the transmission of
macroeconomic signals among companies, particularly in diverse industries. This will
enable the implementation of appropriate state support policies during future crises, which
is crucial given that some companies profited during market turbulence (Shakina and
Barajas 2014).
Considering the existing knowledge on the relationship between profitability and
liquidity, as well as the research method employed, this paper formulated the following
research questions:
1. Is the profitability—liquidity relationship stable in times of crisis?
2. What is the profitability of companies with high and low liquidity?
3. What is the liquidity of companies with high and low profitability?
It is worth mentioning here that this study did not forcefully formulate hypotheses, but
instead focused on posing problems and research questions. Through this approach, many
interesting interpretations were provided, which is acceptable in the fields of economics,
finance, and management sciences. In accordance with the recommendations of scientific
methodologists, it was recognized that hypotheses are not always necessary in these fields
of science. If these conditions were to be met, they would have to be simultaneously new,
general, clear, non-contradictory, and testable. The authors have concluded that meeting all
of these conditions would be somewhat impossible, due in part to the nature of the subject
matter covered, the method employed in this study, as well as its limited framework.

3. Methodology
3.1. Sample
This study uses a dataset spanning from 2019 to 2021, comprising 300 Polish companies
from the wholesale and retail trade sector, both listed and unlisted. These companies
belonged to the G section as per the Polish classification of economic activity (PKD). The
section includes: wholesale and retail sales (i.e., sales not requiring processing) of all types
of goods; provision of services related to the sale of goods; the repair of motor vehicles and
motorcycles. After eliminating outliers and observations with missing data, as is required
by the adopted research methodology, this study utilized 255 companies during the years
2019 and 2020, and 241 companies in 2021. To eliminate univariate outliers, z-scores were
computed and all individual variable observation results that were not within the range of
−3.29 and +3.29 were deleted.
This study used data from the EMIS service provided by an ISI Emerging Markets
Group company. EMIS randomly selected an initial sample of 300 companies as part of its
on-demand data service to minimize the number of observations with missing variables
that are components of profitability and liquidity indicators. It should be noted that the
sample, consisting of companies that reliably report financial data, may have some bias,
although this may not be a significant issue.

3.2. Variables
As far as variables are concerned, accrual profitability was reflected by net profit
margin (NPM), operating return on assets (OROA) and return on equity (ROE). Liquidity
was reflected by current ratio (CURTR), quick ratio (QR) and cash ratio (CR). The operating
Int. J. Financial Stud. 2024, 12, 16 5 of 18

cash flow margin (OCFM) is treated as a cash profitability measure. Generally, cash
efficiency ratios, e.g., OCFM, have characteristics that make them synthetic performance
measures at the intersection of profitability and liquidity that are more resistant to creative
accounting activities than profitability ratios, and as such should be used more often as
the primary synthetic performance measures (Nowicki 2023). Thus, as synthetically as
possible, with the help of seven indicators and taking into account their advantages and
disadvantages, a holistic picture of the company’s profitability and liquidity was created.

3.3. Analytical Approach

This study utilizes the data visualization method of the self-organizing map (SOM),
commonly referred to as Kohonen’s map. SOMs belong to the category of artificial neural
networks, which are trained through unsupervised, competitive learning to acquire a
low-dimensional, discrete representation of the input space derived from the training
samples. Such a representation adopts the form of a map that safeguards the core features
of the input data. Self-organizing maps (SOMs) generally use a two-dimensional grid as
their primary topology, but they can also utilize one-dimensional, multi-dimensional and
irregular grids (Miljković 2017).
This method was chosen primarily for its exploratory nature, which offers the potential
to break free from inconclusive results concerning the subject relationship obtained by more
commonly used methods. Self-Organizing maps provide a straightforward way to gain
insight into the data structure, allowing for the falsification of some of the results presented
in the literature and the development of new hypotheses. To our knowledge, this method
has not been previously applied to either profitability or liquidity, nor has the relationship
between the two been explored. Tonidandel et al. (2018) suggested that “advances in data
science, such as data mining, data visualization, and ML, are extremely well-suited to
address numerous questions in the organizational sciences given the explosion of available
data. Despite these opportunities, few scholars in our field have discussed the specific
ways in which the lens of our science should be brought to bear on the topic of big data and
big data’s reciprocal impact on our science.” This comment also applies, to a lesser extent,
to the discipline of finance.
In this method, each neuron on a map is depicted by an n-dimensional weight vector
w = (w1 , . . ., wn ), where n is the dimension of the input vectors, i.e., the number of variables
used to depict observations. The map’s weights are determined through a learning process
in which the neurons gain comprehension of the fundamental patterns underlying the
data. During this process, a distance measurement is used to compare all data vectors to
all weight vectors. For each input vector, the weights of the nearest neuron are adjusted
to decrease the distance between the input vector and the neuron. Next, the weights of
all neurons located in its neighborhood are also adjusted, but the extent of the change is
directly related to the distance between them on the map. This weight adjustment extends
to all nearby neurons, and the radius of this neighborhood gradually decreases as learning
proceeds using a predefined function. This procedure is repeated until a specified stopping
criterion is met (Du Jardin and Séverin 2012).
Self-organizing maps are well-regarded for their clustering, visualization, and clas-
sification capabilities. In today’s era of ever-expanding datasets, especially in the natural
sciences, the importance of visualization is increasing. Self-organizing maps offer several
appealing attributes in this context: they do not rely on assumptions about data distribu-
tions, can seamlessly handle large datasets, and have proven valuable in numerous practical
applications (Wehrens 2007). One of the most popular neural network architectures is the
self-organizing map (SOM) (Miljković 2017). While there are multiple methods to map a
high-dimensional dataset into two dimensions, with principal component analysis being
the most commonly used, SOM offers numerous advantages (Wehrens 2007).
For this study, we used the Kohonen library (Wehrens and Johannes 2023) in R to create
the maps. Supporting calculations were performed in Excel. A rectangular map consisting
of 16 neurons was used, with 4 per row and 4 per column. While somewhat arbitrary, this
 methods to map a high-dimensional dataset into two dimensions, with principal
component analysis being the most commonly used, SOM offers numerous advantages
(Wehrens 2007).
For this study, we used the Kohonen library (Wehrens and Johannes 2023) in R to
Int. J. Financial Stud. 2024, 12, 16 create the maps. Supporting calculations were performed in Excel. A rectangular6 of map18
consisting of 16 neurons was used, with 4 per row and 4 per column. While somewhat
arbitrary, this configuration corresponds to empirical practices (Cottrell and Rousset
configuration corresponds
1997). The learning to empirical
rate (alpha) is a setpractices (Cottrell
of numbers thatand Rousset
changes 1997). The
gradually fromlearning
0.05 to
rate
0.01(alpha)
over 700is times,
a set ofasnumbers that changes
the complete dataset gradually
is presented from 0.05network
to the to 0.01 over 700The
(rlen). times, as
radius
the complete dataset is presented to the network (rlen). The radius of the
of the neighborhood should begin with a value that covers 2/3 of all unit-to-unit distances, neighborhood
should
changing begin with a value
gradually fromthatthiscovers
value 2/3 of allTo
to zero. unit-to-unit
normalizedistances,
the SOM, changing gradually
the features were
from this value to zero. To normalize the SOM, the features
scaled to unit variance using a common approach (the mean was subtracted from were scaled to unit variance
each
using a common
observation and approach
divided by(the themean
standardwas deviation).
subtracted from each observation and divided
by theAlthough
standard deviation).
it is feasible to identify clusters manually by studying heatmaps and
Although it
generating narrativesis feasible to identify
about distinct clusters
regions manually
on a map,bythis studying
researchheatmaps and generat-
study implemented
ing narratives about distinct regions on a map, this research study
hierarchical clustering provided by the Kohonen library (Wehrens and Johannes implemented hierarchical
2023).
clustering
Ideally, the clusters discovered ought to be adjacent on the map, but this is notthe
provided by the Kohonen library (Wehrens and Johannes 2023). Ideally, clus-
always
ters discovered ought to be adjacent on the map, but this is not always
possible depending on the distribution of underlying variables. The clusters’ quantity waspossible depending
on the distribution
determined by the of underlying
expert technique,variables. The their
observing clusters’ quantity
adjacency and was determined
economic by the
meaning on
expert technique, observing their adjacency
the map. The order of the nodes is shown in Appendix D. and economic meaning on the map. The order
of the nodes is shown in Appendix D.
4. Results
4. Results
4.1.Codes,
4.1. Codes,Node
NodeCounts,
Counts,and
andNeighbor
NeighborDistance
DistancePlots
Plotsin
in2019,
2019,2020,
2020,and
and2021
2021
Theresults
The resultsof
ofthis
thisstudy
studyconcern
concernthethecodes,
codes,node
nodecounts,
counts,and
andneighbor
neighbordistance
distanceplots
plots
for the years 2019–2021. The last two types of plots allow for an in-depth interpretation
for the years 2019–2021. The last two types of plots allow for an in-depth interpretation of
the main area of interest, namely the codes plot, and also permit a visual assessment
of the main area of interest, namely the codes plot, and also permit a visual assessment of of its
quality.
its quality.
Themaps
The mapsfor
for2019
2019are
areshown
shownin inFigure
Figure1.1.

(a) (b) (c)

Figure1.1.Kohonen’s
Figure Kohonen’smap
mapcharacteristics
characteristicsinin2019:
2019:(a)
(a)node
nodeweight
weightvectors
vectorsand
andnode
nodeclusters;
clusters;(b)
(b)node
node
counts; (c) neighbor distance.
counts; (c) neighbor distance.

By examining the weight vectors distributed on the map, individuals can identify
patterns and the distribution of variables and observations. Weight vectors are then
illustrated through individual representations of variable magnitudes for each node using
a “fan” diagram. Afterward, the self-organizing map (SOM) nodes are clustered to group
similar observations with similar “fan” patterns.
As can be seen in Figure 1a, four basic groups of companies can be distinguished in
2019, which are similar in terms of net profit margin (NPM), operating return on assets
(OROA), return on equity (ROE), current ratio (CURTR), quick ratio (QR), cash ratio (CR),
and operating cash flow margin (OCFM). The first group of companies is located in the
most numerous cluster containing 12 nodes marked in green. These are companies with a
diverse but relatively moderate level of ratios. Here, it should be noted that the relative
Int. J. Financial Stud. 2024, 12, 16 7 of 18

level ratio—as the name suggests—is due to comparison with other companies. The
second group of companies—in the nodes marked in red—is characterized by relatively
high profitability (both accrual and cash) as well as relatively high liquidity. The third
cluster, marked in green, contains companies with relatively moderate to high levels of
profitability (both accrual and cash) but at the same time, the relatively lowest levels of
liquidity. The last, blue cluster identified in 2019 contains companies with the relatively
lowest levels of all ratios examined in this study. These four clusters can be labeled with
simplified buzzwords, acting as convenient descriptors for subsequent analysis: “ordinary”
companies, “liquid & profitable” companies, “illiquid & profitable” companies, and “near
bankruptcy” companies.
Figure 1b illustrates the number of observations assigned to each node on the map.
This provides insight into the quality of the map, with an ideal sample distribution being
uniform. The presence of large values in specific areas, such as node 16, may indicate
the benefit of a larger map. However, an excessively large map can result in undesirable
empty nodes.
The plot presented in Figure 1c demonstrates the distance between every node and
its adjacent ones. Areas with a low neighbor distance indicate similar node groups, while
larger distances indicate more dissimilar nodes and natural boundaries between node
clusters. The research sample shows a relatively large distance in node 5, while nodes in
the upper right corner of the plot show a small distance.
Average profitability and liquidity of companies in 2019—based on scaled values of
the variables—in each node are shown in Table 1. For example, the fourth node (“illiquid
& profitable” companies) shares a similar level of OCFM as the fifteenth node (“ordinary”
companies) despite being characterized by incomparably higher levels of NPM.

Table 1. Average values of variables for nodes in 2019.

Node NPM OROA ROE CURTR QR CR OCFM

V1 0.91 0.54 −0.03 3.35 4.24 5.39 1.03
V2 2.28 2.54 0.67 1.31 1.95 2.25 1.65
V3 1.64 1.23 0.53 −0.06 −0.26 −0.32 1.00
V4 6.44 −0.21 −0.12 −1.29 −0.92 −0.38 3.49
V5 0.17 −0.10 −0.26 3.08 2.09 1.04 0.12
V6 −0.87 −0.72 −0.65 0.92 1.45 1.97 −0.53
V7 0.78 1.28 0.39 0.91 0.71 0.05 0.16
V8 0.46 1.43 2.04 −0.59 −0.40 −0.37 0.33
V9 −0.34 −0.36 −0.31 0.48 0.74 0.16 −0.25
V10 −0.42 −0.58 −0.37 −0.56 −0.55 −0.38 0.10
V11 −0.46 −0.55 −0.30 −0.47 −0.52 −0.41 −1.19
V12 −0.23 −0.17 1.91 −0.70 −0.47 −0.19 −0.98
V13 −1.65 −2.47 −5.31 −0.90 −0.72 −0.42 −0.44
V14 −1.58 −1.95 −1.56 −0.41 −0.35 −0.34 −0.66
V15 −0.37 −0.35 −0.22 −0.39 −0.19 −0.18 2.75
V16 −0.01 0.18 0.15 −0.19 −0.36 −0.33 −0.14

In terms of methodology, it is worth noting that the Kohonen library presents the
normalized variables as the default plot. However, in some circumstances, a more practical
and comprehensible visualization would be to present the variables prior to scaling. By
using weight vectors across the map, all dimensions can be viewed simultaneously, which is
unsuitable for a high-dimensional SOM. However, with only seven dimensions, heatmaps
are still valuable. The heatmap enables the visualization of the distribution of a single
variable across the map. Typically, the process of investigating a SOM involves generating
various heatmaps, followed by comparing them to identify notable regions and their
attributes (Lynn 2014). This can be accomplished because the positions of the individual
observations do not vary from one visualization to another.
 scaling. By using weight vectors across the map, all dimensions can be viewed
simultaneously, which is unsuitable for a high-dimensional SOM. However, with only
seven dimensions, heatmaps are still valuable. The heatmap enables the visualization of
the distribution of a single variable across the map. Typically, the process of investigating
Int. J. Financial Stud. 2024, 12, 16 a SOM involves generating various heatmaps, followed by comparing them to identify 8 of 18
notable regions and their attributes (Lynn 2014). This can be accomplished because the
positions of the individual observations do not vary from one visualization to another.
Detailed
Detailed information
information regarding
regarding thethe companies
companies in in 2019
2019 isis displayed
displayed through
through the the
heatmaps
heatmaps presented in Appendix A. Regarding research results, it is worth noting
presented in Appendix A. Regarding research results, it is worth noting that
that
respective nodesareare
respective nodes quite
quite diverse
diverse in comparison
in comparison to 2021,
to 2021, and and2020.
especially especially 2020.
Furthermore,
Furthermore, it is evident that companies exhibiting the highest relative
it is evident that companies exhibiting the highest relative levels of profitability (excluding levels of
profitability
OROA) and (excluding OROA) levels
the lowest relative and the lowest relative
of profitability are levels of profitability the
those demonstrating arelowest
those
demonstrating the lowest
levels of liquidity, levels of liquidity,
while companies while companies
with moderate with
profitability moderate
display profitability
varying degrees
display varying
of liquidity. Thedegrees
clusterofofliquidity.
companies Thewith
cluster
the of companies
highest with on
liquidity, thethe
highest
otherliquidity,
hand, is
on the other hand, is characterized
characterized by moderate profitability. by moderate profitability.
The
The results
results of
of this
this study
study concerning
concerning thethe codes,
codes, node
node counts,
counts, and
and neighbor
neighbor distance
distance
plots for 2020 are shown in Figure 2.

(a) (b) (c)

Figure
Figure 2.
2. Kohonen’s
Kohonen’s map
map characteristics
characteristics in
in 2020:
2020: (a)
(a) node
node weight
weight vectors
vectors and
and node
node clusters;
clusters; (b)
(b) node
node
counts; (c) neighbor distance.
counts; (c) neighbor distance.

As can be
As can be seen
seenininFigure
Figure2a,2a, five
five main
main groups
groups of companies
of companies werewere distinguished
distinguished in
in 2020,
2020, indicating a greater variety of companies than in 2019. The
indicating a greater variety of companies than in 2019. The first group of companies—which first group of
companies—which
can also be described can
as also be described
“ordinary” as “ordinary”
companies—is locatedcompanies—is located cluster.
in the yellow-marked in the
yellow-marked
The second cluster,cluster. The in
marked second
brightcluster, marked in
green, includes bright green,
companies withincludes
relativelycompanies
moderate
with
(OROA)relatively
to highmoderate
(NPM, ROE, (OROA) to highlevels
and OCFM) (NPM, of ROE, and OCFM)
profitability and CRlevels
but atofthe
profitability
same time
and CR butofatCURTR
low levels the sameandtime low levels of companies
QR. Interestingly, CURTR and in QR. Interestingly,
this cluster differ fromcompanies in
“ordinary
this cluster differ
companies” from
only in that“ordinary
they havecompanies”
the relativelyonly in that
highest they
level ofhave thefact,
CR. In relatively highest
this is the only
level of containing
cluster CR. In fact,companies
this is the only
with cluster
such a containing
level of CR.companies with such
The third group a level of CR.
of companies—in
The third group
the nodes markedofincompanies—in the nodes marked
turquoise—is characterized in turquoise—is
by relatively high ROE,characterized
CURTR, and QR by
relatively high ROE,
only. The fourth CURTR,
cluster, markedandinQRred,only. The companies
contains fourth cluster,
withmarked
relativelyin high
red, contains
levels of
NPM and OCFM only. The fifth cluster, marked in blue, consists of companies which are
unique, in that they are the only ones with the relatively highest levels of profitability (both
accrual and cash) and the lowest levels of liquidity.
The plot in Figure 2b indicates large values in the third and eighth nodes. The plot in
Figure 2c indicates that there is a relatively large distance in the thirteenth node.
Average profitability and liquidity of companies in 2020—based on scaled values of
the variables—in each node are shown in Table 2.
Int. J. Financial Stud. 2024, 12, 16 9 of 18

Table 2. Average values of variables for nodes in 2020.

Node NPM OROA ROE CURTR QR CR OCFM

V1 −0.27 −1.75 −14.50 −0.94 −0.64 −0.62 −0.25
V2 −0.77 −3.44 −1.40 −0.67 −0.46 −0.44 −0.23
V3 0.04 −0.44 0.06 −0.41 −0.36 −0.54 0.04
V4 −0.03 −1.19 0.01 −0.51 −0.44 −0.36 0.04
V5 −0.02 −0.98 0.03 −0.14 −0.13 0.99 0.00
V6 0.06 −0.30 0.06 −0.12 −0.05 0.35 0.18
V7 0.10 0.32 0.08 −0.05 −0.17 −0.35 0.07
V8 0.06 −0.05 0.09 −0.58 −0.50 −0.47 0.05
V9 0.16 0.59 0.07 2.42 2.05 3.57 0.14
V10 0.05 −0.31 0.05 0.78 0.72 0.12 −0.06
V11 0.09 0.45 0.08 0.19 0.16 0.93 0.07
V12 0.15 1.21 0.15 −0.39 −0.28 −0.36 0.18
V13 −15.60 −0.88 0.04 9.94 11.52 −0.63 −15.73
V14 0.06 −0.11 0.06 1.01 1.11 1.99 −0.01
V15 0.15 1.06 0.09 0.79 0.55 −0.06 0.12
V16 0.45 4.51 0.70 0.04 0.10 −0.09 0.11

In-depth information about the companies in 2020 is shown in the heatmaps in Ap-
pendix B. Regarding research results, it is worth noting that respective nodes are very
similar except from CR, and—to a lesser extent—OROA. What is particularly interesting is
that in 2020, unlike in 2019 and 2020, the lowest value of the NPM map is in a completely
different place than the lowest values of the OROA and ROE maps, despite the fact that they
are all ratios of accrual profitability. As far as the relationships of the highest profitability
as well as the lowest profitability with the lowest liquidity are concerned, in contrast to
2019 and 2021, there are no clear patterns in 2020.
Int. J. Financial Stud. 2024, 12, x FOR PEER REVIEW 10 of 19
The results of this study concerning the codes, node counts, and neighbor distance
plots for 2021 are shown in Figure 3.

(a) (b) (c)

Figure 3.
Figure 3. Kohonen’s
Kohonen’s map
map characteristics
characteristics in
in 2021:
2021: (a)
(a) node
node weight
weight vectors
vectors and
and node
node clusters;
clusters; (b)
(b) node
node
counts; (c) neighbor distance.
counts; (c) neighbor distance.

Six main
Six main clusters
clusters of
of companies
companies may
may bebe identified
identified in
in 2021,
2021, as
as shown
shown inin Figure
Figure 3a.
3a. This
This
indicates an
indicates an increase
increase in
in diversity
diversity among
among companies.
companies. There
There are
are two
two reasons
reasons for
for this.
this. Firstly,
Firstly,
the research
the research methodology
methodology used
used aa baseline
baseline of
of 2019
2019 and
and outliers
outliers were
were only
only removed
removed in in this
this
year.
year. Secondly, the aftermath
aftermath of the 2020 crisis may have caused companies
the 2020 crisis may have caused companies to divergeto diverge
further. The first, the most
further. most numerous
numerous group
group ofof companies,
companies, which
which areare characterized
characterized by aa
relatively high level of OCFM and relatively moderate levels of other ratios, is marked in
orange. The second cluster, marked in bright green, includes companies with a relatively
high level of OCFM, and relatively low levels of other ratios. The third cluster, marked in
red, is characterized by relatively moderate to high levels of profitability, and the
relatively lowest levels of liquidity. The fourth, light green cluster, is characterized by a
Int. J. Financial Stud. 2024, 12, 16 10 of 18

relatively high level of OCFM and relatively moderate levels of other ratios, is marked in
orange. The second cluster, marked in bright green, includes companies with a relatively
high level of OCFM, and relatively low levels of other ratios. The third cluster, marked in
red, is characterized by relatively moderate to high levels of profitability, and the relatively
lowest levels of liquidity. The fourth, light green cluster, is characterized by a relatively
high level of NPM only. The fifth cluster, marked in light blue, is the only one that contains
companies with relatively high liquidity readings. The last cluster, marked in dark blue,
contains companies with balanced and at the same time moderate levels of all ratios.
The plot in Figure 3b indicates large values in a few nodes, mainly the eleventh and
twelfth. The plot in Figure 3c indicates that there is a relatively large distance in the
thirteenth node—the one associated with relatively high levels of liquidity.
Average profitability and liquidity of companies—based on scaled values of the
variables—in each node are shown in Table 3.

Table 3. Average values of variables for nodes in 2021.

Node NPM OROA ROE CURTR QR CR OCFM

V1 1.55 3.55 2.29 0.64 0.30 −0.10 0.18
V2 0.68 1.24 0.51 0.19 −0.04 −0.34 −0.15
V3 −0.15 0.26 2.96 −0.82 −0.44 −0.26 0.11
V4 −0.11 0.08 0.75 −0.68 −0.53 −0.39 0.08
V5 4.89 −0.98 −0.12 0.21 0.86 −0.51 −5.78
V6 3.06 0.27 −0.06 0.07 0.22 −0.17 1.23
V7 −0.07 0.08 −0.20 −0.08 0.02 0.37 0.29
V8 0.12 −0.49 −0.06 −0.70 −0.37 −0.39 2.51
V9 0.80 0.92 −0.04 0.77 0.91 1.56 0.66
V10 −0.15 −0.32 −0.56 1.08 0.72 0.17 −0.20
V11 −0.22 −0.11 0.01 −0.31 −0.40 −0.42 −0.47
V12 −0.52 −0.81 −0.66 −0.67 −0.60 −0.34 −0.08
V13 0.84 0.32 −0.36 6.79 8.58 9.67 0.66
V14 0.29 0.08 −0.50 2.75 2.67 2.74 0.06
V15 −1.24 −1.49 −1.44 −0.43 −0.26 0.26 −0.41
V16 −2.59 −1.66 −1.60 −0.29 −0.08 −0.28 3.74

In-depth information about the companies in 2021 is shown on the heatmaps in

Appendix C. Regarding the research results, it is worth noting that respective nodes are
very similar in terms of liquidity, and less in terms of profitability (accrual and cash). Of
particular interest may be the companies grouped in node 5, which are characterized by
the relatively lowest OCFM, and at the same time the relatively highest NPM. This is quite
unusual, even more so considering 2019 and 2020, where the same node, in this case node
4, had the relatively highest levels of OCFM and NPM. The similar strange regularity in
2021 can be observed in node 16, where the highest indications for OCFM are the lowest
for the other profitability ratios. Moreover, it can be seen that companies with the highest
profitability (accrual and cash) as well as the lowest profitability are those with the lowest
or moderate liquidity, while companies with moderate profitability have various levels of
liquidity, although the conclusions drawn from Table 3 may be somewhat more nuanced
than those drawn from the heatmaps in Appendix C, as the heatmaps show only relatively
large differences in the levels of the indicators. The cluster of companies with the highest
liquidity, on the other hand, is characterized by moderate profitability.

4.2. The Trajectory of High Profitability—High Liquidity Companies over 2019–2021

The results of this study concern the stability of the profitability—liquidity relationship
over time. The stability—using Kohonen’s maps—is shown, when the same companies are
located in nodes with similar characteristics of the variables over the years studied. The
most interesting relationship, i.e., the high profitability—high liquidity relationship in the
pre-crisis, during the crisis, and in the post-crisis periods is presented in Figure 4.
 The results of this study concern the stability of the profitability—liquidity
relationship over time. The stability—using Kohonen’s maps—is shown, when the same
companies are located in nodes with similar characteristics of the variables over the years
studied. The most interesting relationship, i.e., the high profitability—high liquidity
Int. J. Financial Stud. 2024, 12, 16
relationship in the pre-crisis, during the crisis, and in the post-crisis periods is presented
11 of 18
in Figure 4.

Figure 4. The high profitability—high liquidity relationship in the pre-crisis, during the crisis, and in
Figure 4. The high
the post-crisis profitability—high liquidity relationship in the pre-crisis, during the crisis, and
periods.
in the post-crisis periods.
As can be seen in Figure 4, in 2019, the nodes with relatively high levels of liquidity
and profitability are nodes 1 and 2 (red color). These nodes included 10 companies, of
which 9 are shown due to the lack of data in 2020 for the 10th company. As many as
seven of these nine companies were in 2020 at nodes 9 and 14 (green color), i.e., the nodes
with the most balanced levels of profitability and liquidity ratios in that year. This means
that these companies are mostly following the same path. And although the liquidity
of these companies in 2020 dropped significantly relative to 2019, they are still the most
balanced companies, i.e., those maintaining both profitability and liquidity. Interestingly,
these companies were characterized by the relatively highest CR. At the same time, none
of the companies with the relatively highest liquidity and profitability in 2019 were in the
node of companies with the relatively highest liquidity and lowest profitability in 2020, i.e.,
the thirteenth node marked in light blue. In 2021, seven out of the nine initially analyzed
companies were in balanced nodes, specifically nodes nine, thirteen, and fourteen, although
in six cases, the balance meant relatively moderate levels of liquidity and profitability. Gen-
erally, it can be noted that companies that had a balanced profitability–liquidity relationship
in the pre-crisis period maintained this relationship during the crisis and in the post-crisis,
although the level of these ratios in the analyzed research sample declined.

5. Discussion, Limitations and Future Research

This study utilizes the self-organizing map (SOM) data visualization technique, a
method not previously explored within the field it addresses. This approach allows for
conclusions that reconcile the partially conflicting research results found in the existing lit-
erature. Consequently, this study enriches the academic discourse and carries implications
for practical application in business.
This study’s results, which cover three different time frames, offer a new perspective
on the issue identified by Cheong et al. (2021) who claimed that “although there has been
a myriad of research works explaining the effect of the crisis on firm performance, no
study has attempted to examine the determinants of both accounting and market-based
measures of firm profitability over the pre-crisis, crisis and post-crisis periods.” Still, in
the previous research, liquidity appears to have a limited impact as it only influences the
return on equity in the period following the crisis (Cheong et al. 2021). In a similar vein,
Int. J. Financial Stud. 2024, 12, 16 12 of 18

Nanda and Panda (2018) found that in the post-crisis period, size and liquidity were signifi-
cantly and positively related, whereas leverage was significantly and negatively related to
profitability. In detail, their study confirmed that Indian manufacturing firms can enhance
their profitability by maintaining sufficient liquidity. Firms may incur some short-term
costs for maintaining liquidity, but in the long run, it has a positive impact on profitability.
Accordingly, their study supports the hypothesis that an increase in firm size and liquidity
enhances profitability, while leverage discourages profitability. During the crisis, both debt
and the availability of cash were important elements (Notta and Vlachvei 2014). Moreover,
while the agency theory of free cash flow argues that accumulating corporate cash can
lead managers to invest in projects that decrease value, maintaining high cash reserves is
advised during a crisis to ensure corporate liquidity (Jensen 1986). In terms of the pre-crisis
period, the debt–equity ratio and liquidity play significant roles and have opposite effects
on corporate profits, one negatively and the other positively (Nanda and Panda 2018). In
particular, companies with pre-existing financial adaptability possess a higher capacity to
seize investment prospects, are far less dependent on internal funds for investment, and
outperform less adaptable firms during the crisis (Arslan-Ayaydin et al. 2014).
Following the established causal link where liquidity influences profitability, this
study concludes that during non-crisis periods, low liquidity is a necessary but insufficient
condition for the exceptionally high profitability. Furthermore, given that the highest
levels of liquidity were recorded for companies with moderate profitability, the relationship
between profitability and liquidity may deny the shape of a line of an inverted “U” letter
(Adamczyk and Waśniewski 2018; Gentry 1976; Kowerski 2016) and suggest—at least at
first glance—the shape of an area of a “D” letter. These companies with high liquidity and
moderate profitability are protected against bankruptcy due to their liquidity. However,
they face limitations on significant performance for the same reason.
In summary, for the majority of companies during non-crisis periods, the relationship
between profitability and liquidity aligns with the midpoint of Gentry’s curve (Gentry 1976).
This signifies that profitability is not reliant on liquidity, suggesting an absence of a clear
relationship. Differing findings in studies (showing positive or negative relationships)
might stem from intra-industry variations. This echoes Hawawini et al. (2003) who
indicated that exceptional firms significantly differ in performance from others in the same
industry, where firm-specific assets outweigh industry factors. This implies that liquidity’s
impact on profitability may only be significant for exceptional companies, aligning with
Awad et al. (2013) who found negligible statistical impacts of average liquidity levels on
profitability. Still, the interpretation could go further as according to Hirigoyen (1985),
profitability and liquidity not only result from each other but also impose limitations on
one another.
Considering that an interesting phenomenon emerged in the post-crisis period of
2021, with companies having their relatively lowest levels of OCFM while also having their
relatively highest levels of NPM, it is worth reaching out to Nowicki (2023) who argued
that cash efficiency ratios are more resilient to creative accounting practices than accrual
profitability ratios and, as such, should be used more frequently as primary synthetic
efficiency measures, which currently seem to be dominated by accrual profitability ratios.
In general, we can say that different ratios react in different ways and at different speeds.
In this context, an interesting study was undertaken by Nanda and Panda (2018) claiming
that NPM reacts to changes in debt more swiftly than ROA, and in detail, that the initial
adverse effect on ROA caused by a substantial debt change is relatively smaller compared
to NPM, and it diminishes more rapidly than NPM.
This study has limitations in sample selection, year selection, and the method used.
Specifically, it is biased towards a particular country and industry, which reduces the
generalizability of its findings. Additionally, using only one year (2019) as a baseline is a
limitation. Including more years would be valuable, especially since crises have different
characteristics. Furthermore, there are various ways of defining pre- and post-crisis periods,
as well as the crises themselves. Official data from state statistical offices are used to
Int. J. Financial Stud. 2024, 12, 16 13 of 18

determine when the crisis has ended, as is common practice in most studies. However,
it is important to note that the consequences of the crisis can be nuanced and prolonged,
and the recovery process can be lengthy. Therefore, 2020 can be considered the central
crisis period. It is important to note that the post-crisis period does not necessarily mean
that it is a completely positive time for companies. Finally, although SOM has advantages
with its intuitive visual presentation, it does not provide statistically validated numbers
to prove the results. Another limitation of this study is the comparability with previous
studies due to its unique methodology. In future research, it may be necessary to combine
the advantages of SOM clustering with classical regression analysis or classification.
The research conducted can be continued in several directions. First, the relationship
should be studied for selected companies, such as young ones, or listed ones, etc. In this
context, Varum and Rocha (2013) argue that small and medium-sized enterprises act as
stabilizers during economic downturns, while the performance of larger firms appears
to decline more rapidly. Small firms are more agile and better equipped to adapt to
changes in their external environment, thereby experiencing less hardship during economic
adversities. Second, incorporating moderators into the research is advisable. For example,
Jaworski and Czerwonka (2021) suggest that various factors can moderate the strength
and direction of this correlation, linked to macroeconomic and institutional circumstances.
Furthermore, exploring distinct industry profiles could offer valuable insights. For instance,
Chambers and Cifter (2022) highlighted that “across a broad analysis of seven U.S. industrial
sectors, Jose et al. (1996) found that proactive management of liquidity is linked to increased
profitability in several industries, although not universally across all sectors.”

6. Conclusions
The aim of this study was to verify the stability of the profitability–liquidity relation-
ship over time, as well as to determine this relationship in terms of its level and structure.
This study’s primary findings are as follows.
Firstly, it can be generally noted that companies with a balanced profitability–liquidity
relationship in the pre-crisis period maintain this relationship during the crisis and in the
post-crisis periods, although the levels of profitability and liquidity ratios in the analyzed
research sample declined.
Secondly, the results of this study show that companies in the clusters with the
relatively highest and lowest profitability have the relatively lowest or moderate liquidity
both before and after the crisis. Thus, it can be noted that every company with very
high profitability is characterized by low liquidity, which of course does not mean that
every company with low liquidity is characterized by very high profitability. Moreover,
companies with relatively moderate profitability have different levels of liquidity. On
the other hand, the group of companies with the relative highest liquidity tends to have
relatively moderate profitability. However, these patterns are not evident during a crisis.
Thirdly, an interesting phenomenon emerged in the post-crisis period of 2021, with
companies having their relatively lowest levels of OCFM while also having their relatively
highest levels of NPM. This phenomenon is particularly striking when compared to 2019
and 2020, when the same companies had the highest levels of both OCFM and NPM
simultaneously. This unusual pattern is again evident in 2021 when examining companies
with the relatively highest OCFM but the relatively lowest accrual profitability ratios. The
years 2019 and 2020 are characterized by a “normal” convergence of the highest levels for
OCFM and most profitability ratios.

Author Contributions: Conceptualization, P.R., J.N. and D.S.; methodology, P.R., J.N. and D.S.;
validation, P.R. and D.S.; formal analysis, P.R.; investigation, P.R.; resources, P.R., J.N. and D.S.; data
curation, P.R., J.N. and D.S.; writing—original draft preparation, P.R., J.N. and D.S.; writing—review
and editing, P.R. and D.S.; funding acquisition, D.S. All authors have read and agreed to the published
version of the manuscript.
Int. J. Financial Stud. 2024, 12, 16 14 of 18

Funding: This publication was financed from the state budget under the program of the Min-
ister of Education and Science (Poland) under the name “Science for Society”. Project number:
NdS/543640/2021/2022, amount of co-financing: 155.125 PLN, total value of the project: 699.200 PLN.
Informed Consent Statement: Not applicable.
Data Availability Statement: Data are available on the paid platform EMIS provided by an ISI
Emerging Markets Group company.
Int. J. Financial Stud. 2024, 12, x FORConflicts of Interest: The authors declare no conflicts of interest.
PEER REVIEW 15 of 19

Appendix A. Heatmaps for 2019

Appendix B. Heatmaps for 2020

Int. J. Financial Stud. 2024, 12, 16 15 of 18
Int. J. Financial Stud. 2024, 12, x FOR PEER REVIEW 16 of 19

Appendix B. Heatmaps for 2020

Appendix C. Heatmaps for 2021

Int.
Int.J.J.Financial
FinancialStud. 2024,12,
Stud.2024, 12,16
x FOR PEER REVIEW 16 of
17 of 18
19
Int. J. Financial Stud. 2024, 12, x FOR PEER REVIEW 17 of 19

Appendix C. Heatmaps for 2021

Appendix D. The Order of the Nodes

Appendix
AppendixD.
D.The
TheOrder
Orderofofthe
theNodes
Nodes

References
References
(Adamczyk and Waśniewski 2018) Adamczyk, Adam, and Piotr Waśniewski. 2018. Is Excess Liquidity Real? The Analysis of
(Adamczyk and Waśniewski
Relationship 2018) Adamczyk,
between Liquidity Adam, and
and Profitability. PiotrNaukowe
Zeszyty Waśniewski. 2018. Is Excess
Uniwersytetu Liquidity
Szczecińskiego Real? Rynki
Finanse The Analysis of
Finansowe
Relationship 92:
Ubezpieczenia between Liquidity and Profitability. Zeszyty Naukowe Uniwersytetu Szczecińskiego Finanse Rynki Finansowe
11–21. https://doi.org/10.18276/FRFU.2018.92-01.
Ubezpieczenia 92: 11–21. https://doi.org/10.18276/FRFU.2018.92-01.
Int. J. Financial Stud. 2024, 12, 16 17 of 18

References
Adamczyk, Adam, and Piotr Waśniewski. 2018. Is Excess Liquidity Real? The Analysis of Relationship between Liquidity and
Profitability. Zeszyty Naukowe Uniwersytetu Szczecińskiego Finanse Rynki Finansowe Ubezpieczenia 92: 11–21. [CrossRef]
Adelopo, Ismail, Robert Lloydking, and Venancio Tauringana. 2018. Determinants of Bank Profitability before, during, and after the
Financial Crisis. International Journal of Managerial Finance 14: 378–98. [CrossRef]
Arslan-Ayaydin, Özgür, Chris Florackis, and Aydin Ozkan. 2014. Financial Flexibility, Corporate Investment and Performance:
Evidence from Financial Crises. Review of Quantitative Finance and Accounting 42: 211–50. [CrossRef]
Awad, Ibrahim, Fahema Jayya, Ibrahim Awad, and Fahema Jayya. 2013. Working Capital Management, Liquidity and Profitability of
the Manufacturing Sector in Palestine: Panel Co-Integration and Causality. Modern Economy 4: 662–71. [CrossRef]
Byczkowska, Magdalena, and Andrzej Kuciński. 2018. Ocena Płynności Finansowej Przedsi˛ebiorstw Na Podstawie Analizy Rachunku
Przepływów Pieni˛eżnych. Zeszyty Naukowe Małopolskiej Wyższej Szkoły Ekonomicznej w Tarnowie 38: 85–100. [CrossRef]
Chambers, Nurgul, and Atilla Cifter. 2022. Working Capital Management and Firm Performance in the Hospitality and Tourism
Industry. International Journal of Hospitality Management 102: 103144. [CrossRef]
Cheong, Carol, Huy Viet Hoang, and David Mcmillan. 2021. Macroeconomic Factors or Firm-Specific Factors? An Examination of the
Impact on Corporate Profitability before, during and after the Global Financial Crisis. Cogent Economics & Finance 9: 1959703.
[CrossRef]
Chow, Yee Peng, Junaina Muhammad, A. N. Bany-Ariffin, and Fan Fah Cheng. 2018. Macroeconomic Uncertainty, Corporate
Governance and Corporate Capital Structure. International Journal of Managerial Finance 14: 301–21. [CrossRef]
Cottrell, Marie, and Patrick Rousset. 1997. The Kohonen Algorithm: A Powerful Tool for Analysing and Representing Multidimensional
Quantitative and Qualitative Data. In Biological and Artificial Computation: From Neuroscience to Technology. Lecture Notes in
Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Berlin and
Heidelberg: Springer, vol. 1240 LNCS, pp. 861–71. [CrossRef]
Du Jardin, Philippe, and Eric Séverin. 2012. Forecasting Financial Failure Using a Kohonen Map: A Comparative Study to Improve
Model Stability over Time. European Journal of Operational Research 221: 379–96. [CrossRef]
Eljelly, Abuzar M. A. 2004. Liquidity—Profitability Tradeoff: An Empirical Investigation in an Emerging Market. International Journal of
Commerce and Management 14: 48–61. [CrossRef]
Gentry, James. 1976. Management Perceptions of the Working Capital Process. Champaign: College of Commerce and Business Ad-
ministration, University of Illinois at Urbana. Available online: https://core.ac.uk/download/pdf/4835142.pdf (accessed on
10 December 2023).
Hawawini, Gabriel, Venkat Subramanian, and Paul Verdin. 2003. Is Performance Driven by Industry—Or Firm-Specific Factors? A
New Look at the Evidence. Strategic Management Journal 24: 1–16. [CrossRef]
Hirigoyen, Gerard. 1985. Rentabilité et Solvabilité. Direction et Gestion, 3, 13–26.—References—Scientific Research Publishing.
Available online: https://www.scirp.org/(S(351jmbntvnsjt1aadkozje))/reference/referencespapers.aspx?referenceid=2170423
(accessed on 5 December 2023).
Hirsch, Stefan, Jan Schiefer, Adelina Gschwandtner, and Monika Hartmann. 2014. The Determinants of Firm Profitability Differences
in EU Food Processing. Journal of Agricultural Economics 65: 703–21. [CrossRef]
Jaworski, Jacek, and Leszek Czerwonka. 2021. Meta-Study on the Relationship between Profitability and Liquidity of Enterprises in
Macroeconomic and Institutional Environment. Decision 48: 233–46. [CrossRef]
Jaworski, Jacek, Leszek Czerwonka, and Magdalena Madra-Sawicka.
˛ 2018. Zależność Mi˛edzy Rentownościa˛ a Płynnościa˛ Finansowa˛
w Sektorze Przetwórstwa Spożywczego w Polsce. Roczniki Naukowe Stowarzyszenia Ekonomistów Rolnictwa i Agrobiznesu 20: 58–63.
[CrossRef]
Jensen, Michael. 1986. Agency Costs of Free Cash Flow, Corporate Finance, and Takeovers. American Economic Review 76: 323–29.
Jose, Manuel L., Carol Lancaster, and Jerry L. Stevens. 1996. Corporate returns and cash conversion cycles. Journal of Economics and
Finance 20: 33–46. [CrossRef]
Kowerski, Mieczysław. 2016. Zależność Mi˛edzy Rentownościa˛ a Płynnościa˛ Finansowa˛ Ma Kształt Odwróconego U. Prace Naukowe
Uniwersytetu Ekonomicznego We Wrocławiu 440: 338–48. [CrossRef]
Lee, Chien Chiang, Mei Ping Chen, and Shao Lin Ning. 2017. Why Did Some Firms Perform Better in the Global Financial Crisis?
Economic Research-Ekonomska Istraživanja 30: 1339–66. [CrossRef]
Lynn, Shanel. 2014. Self-Organising Maps for Customer Segmentation Using R. Available online: https://www.shanelynn.ie/self-
organising-maps-for-customer-segmentation-using-r/ (accessed on 5 December 2023).
Maçãs Nunes, Paulo J., Zélia M. Serrasqueiro, and Tiago N. Sequeira. 2009. Profitability in Portuguese Service Industries: A Panel Data
Approach. The Service Industries Journal 29: 693–707. [CrossRef]
Makhija, Mona. 2003. Comparing the Resource-Based and Market-Based Views of the Firm: Empirical Evidence from Czech
Privatization. Strategic Management Journal 24: 433–51. [CrossRef]
Miljković, Dubravko. 2017. Brief Review of Self-Organizing Maps, 1252–57. Available online: https://ieeexplore.ieee.org/document/
7973581 (accessed on 1 December 2023).
Nanda, Swagatika, and Ajaya Kumar Panda. 2018. The Determinants of Corporate Profitability: An Investigation of Indian Manufac-
turing Firms. International Journal of Emerging Markets 13: 66–86. [CrossRef]
Int. J. Financial Stud. 2024, 12, 16 18 of 18

Notta, Ourania, and Aspasia Vlachvei. 2014. The Impact of Financial Crisis on Firm Performance in Case of Greek Food Manufacturing
Firms. Procedia Economics and Finance 14: 454–60. [CrossRef]
Nowicki, Jarosław. 2023. Rentowność czy wydajność gotówkowa? Syntetyczny pomiar efektywności w kontekście działań z zakresu
kreatywnej rachunkowości. In Analiza ekonomiczna w warunkach niepewności. Edited by Cezary Kochalski. Poznań: Wydawnictwo
Uniwersytetu Ekonomicznego w Poznaniu, pp. 102–17. [CrossRef]
Nunes, Paulo Maçãs, Ana Viveiros, and Zélia Serrasqueiro. 2012. Are the Determinants of Young SME Profitability Different? Empirical
Evidence Using Dynamic Estimators. Journal of Business Economics and Management 13: 443–70. [CrossRef]
Pervan, Maja, Ivica Pervan, and Marijana Ćurak. 2019. Determinants of Firm Profitability in the Croatian Manufacturing Industry:
Evidence from Dynamic Panel Analysis. Economic Research-Ekonomska Istrazivanja 32: 968–81. [CrossRef]
Ratajczak, Piotr. 2023. Holistyczna koncepcja procesu bankructwa przedsi˛ebiorstwa. In Analiza ekonomiczna w warunkach niepewności.
Edited by Cezary Kochalski. Poznań: Wydawnictwo Uniwersytetu Ekonomicznego w Poznaniu, pp. 118–34. [CrossRef]
Shakina, Elena, and Angel Barajas. 2014. The Dynamics of Intellectual Resources during the Economic Crisis. Economic Research-
Ekonomska Istrazivanja 27: 861–81. [CrossRef]
Sur, Debasish, and Kaushik Chakraborty. 2011. Evaluating Relationship of Working Capital and Profitability: A Study of Select
Multinational Companies in the Indian Pharmaceutical Sector. The IUP Journal of Management Research: IJMR 10: 7.
Tonidandel, Scott, Eden B. King, and Jose M. Cortina. 2018. Big Data Methods: Leveraging Modern Data Analytic Techniques to Build
Organizational Science. Organizational Research Methods 21: 525–47. [CrossRef]
Varum, Celeste Amorim, and Vera Catarina Rocha. 2013. Employment and SMEs during Crises. Small Business Economics 40: 9–25.
[CrossRef]
Vieira, Renato Schwambach. 2010. The Relationship between Liquidity and Profitability. An Exploratory Study of Airline Companies
between 2005 and 2008. Unpublished Master’s thesis, Umeå University, Umeå, Sweden. Available online: http://www.diva-
portal.org/smash/get/diva2:409560/FULLTEXT01.pdf (accessed on 7 December 2023).
Wehrens, Ron. 2007. Self- and Super-organizing Maps in R: The Kohonen Package. JSS Journal of Statistical Software 21: 1–19. Available
online: http://www.jstatsoft.org/ (accessed on 7 December 2023).
Wehrens, Ron, and Kruisselbrink Johannes. 2023. Package ‘kohonen’ Title Supervised and Unsupervised Self-Organising Maps.
Available online: https://cran.r-project.org/web/packages/kohonen/kohonen.pdf (accessed on 2 December 2023).

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.