Binary logistic regression can be used to asses whether the company has gone bankrupt or not.
variables used:-
R9-current ratio R21-asset/debts R24-wcfo/debt
Rcode-
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
##
## Call:
## glm(formula = Bankrupt ~ R9 + R21 + R24, family = binomial)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.20530 -0.73295 -0.04158 0.41851 2.31406
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -5.9044 1.4741 -4.005 6.19e-05 ***
## R9 1.1881 0.4764 2.494 0.0126 *
## R21 1.9844 0.9081 2.185 0.0289 *
## R24 4.6583 2.1745 2.142 0.0322 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 182.99 on 131 degrees of freedom
## Residual deviance: 105.12 on 128 degrees of freedom
## AIC: 113.12
##
## Number of Fisher Scoring iterations: 7
AIC: 113.12
PVALUES- R9-0.0126 R21-0.0289 R24-0.0322
Bankruptcy signals are indicators or signs that a company may be experiencing financial distress and may be at risk of declaring bankruptcy. These signals can be used by investors, creditors, and other stakeholders to assess the financial health and viability of a company.
Some common bankruptcy signals include: 1)poor current ratio 2)low asset to debt coverage 3)cash flow from operations to debt coverage
It is important to note that none of these signals alone can definitively predict bankruptcy, and companies can experience financial distress for a variety of reasons. However, paying attention to these signals can help stakeholders make more informed decisions about their investments or business relationships with the company in question.
attach(Bankruptdata)
## The following objects are masked from Bankruptdata4:
##
## Bankrupt, R1, R10, R11, R12, R13, R14, R15, R16, R17, R18, R19, R2,
## R20, R21, R22, R23, R24, R3, R4, R5, R6, R7, R8, R9, YR
bankrupts1<-split(Bankruptdata,Bankrupt)
crbind <- cbind(bankrupts1$No$R9,bankrupts1$Yes$R9)
boxplot(crbind,ylim = c(0.2,5),names =c("NOT Bankrupt","Bankrupt"),xlab = "R9 - Current ratio")
adbind <- cbind(bankrupts1$No$R21,bankrupts1$Yes$R21)
boxplot(adbind,ylim = c(0.5,4.5),names =c("NOT Bankrupt","Bankrupt"),xlab = "R21 - asset/debt")
wdbind <- cbind(bankrupts1$No$R24,bankrupts1$Yes$R24)
boxplot(wdbind,ylim = c(-0.7,1),names =c("NOT Bankrupt","Bankrupt"), xlab = "R24-wcfo/debts")
Model: After performing binary logistic regression
on the variables r9,r21,r24 we can infer that these three ratios are key
indicators to predictiong the future bankruptcy probability of the
company.
After observing the p values of these variables it can be noted that the current ratio is the most important variable because of the following reasons:-
Liquidity: The current ratio is a measure of a company’s liquidity, which refers to its ability to convert assets into cash to meet short-term obligations. If a company has a low current ratio, it may not have enough liquidity to meet its obligations and may be at risk of bankruptcy.
Solvency: The current ratio is also a measure of a company’s solvency, which refers to its ability to meet long-term obligations. A company with a low current ratio may not have the financial resources to meet its long-term obligations, which can increase the risk of bankruptcy.
Creditworthiness: The current ratio is often used by lenders to assess a company’s creditworthiness. If a company has a low current ratio, it may have difficulty obtaining credit, which can limit its ability to operate and grow.
Industry Standards: The current ratio can also be compared to industry standards to assess a company’s financial health. If a company’s current ratio is significantly lower than the industry average, it may be an indication of financial distress.