A proposed multidimensional model for predicting financial distress: an empirical study on Egyptian listed firms

Future Business Journal

Table 4 Logistic regression results

FD	Model 1 (Financial Variables)	Model 2 (CGI)	Model 3 (Market)	Model 4 (Macroeconomic)	Model 5 (Restricted/nested)	Model 6 (Unrestricted/non-nested)
WCTA	− 2.093**(0.877)				− 1.975*(1.034)	− 2.323**(1.054)
RETA	0.765*** (0.242)
EBITTA	− 12.328*** (2.892)				− 17.411***(3.852)	− 24.904***(4.538)
SalesTA	− 2.118*** (0.657)				− 2.072***(0.6863)	− 2.062***(0.634)
MVETL	0.092** (0.037)				0.183***(0.070)	0.185***(0.070)
CGI		− 3.318***(1.213)
lnMcap			− 0.546***(0.114)		− 0.311**(0.161)
IR				− 0.446**(0.222)	− 0.538***(0.129)
GDP				0.204***(0.075)	0.462***(0.102)	0.736***(0.155)
CPI				0.191(0.159)
COVID-19						5.761***(1.277)
Constant	− 1.122*** (0.280)	− 0.002 (0.741)	9.199***(2.312)	− 0.822(0.778)	4.994*(3.453)	− 15.513***(3.147)
Observations	458	463	458	472	458	458
Log likelihood	− 114.4788	− 160.955	− 150.421	− 142.847	− 80.824	− 78.549
LR \({X}^{2}\)	99.26***	7.530***	27.370***	45.910***	166.57***	171.12***
Pseudo \({R}^{2}\)	0.3024	0.023	0.083	0.1384	0.508	0.521
McKelvey and Zavoina’s \({R}^{2}\)					0.965	0.984
Cragg and Uhler’s \({R}^{2}\)					0.596	0.609
Pearson \({X}^{2}\)					178.33	188.84
Hosmer–Lemeshow \({X}^{2}\)					2.39	4.88
LR test—nested model \({X}^{2}\)					7.21***
Wald test					5.84**
AIC					177.648	171.098
BIC					210.662	199.986
Area under ROC curve					0.946	0.955
Accuracy ratio (AR)					0.892	0.911

Model 1: Model with financial variables; Model 2: Model with corporate governance index score; Model 3: Model with market indicators; Model 4: Model with macroeconomic variables; Model 5: Multidimensional model; Model 6: Multidimensional model after controlling for COVID-19 effects FD: Financial Distress; WCTA: Working capital to total assets; RETA: Retained earnings to total assets; EBITTA: Earnings before interest and taxes to total assets; MVETL: Market value of equity to total liabilities; SalesTA: Sales to total assets; CGI: Corporate governance index score; lnMcap: Natural logarithm of market capitalization; IR: Interest rate; GDP: GDP growth rate; CPI: Consumer price index; COVID-19: Dummy variable for COVID-19 years. LR \({X}^{2}:\) Pregibon [64] likelihood ratio test for model significance Pseudo \({R}^{2}\), McKelvey and Zavoina’s \({R}^{2}\), Cragg and Uhler’s \({R}^{2}:\) measures of goodness-of-fit tests AIC and BIC: The Akaike information criteria and Bayesian information criteria for model selection Area under ROC curve: measure of classification accuracy,AR: Accuracy ratio = 2 × (AUC − 0.5) [42]. Figures in brackets are standard errors, ***P value < 0.01; **P value < 0.05; *p value < 0.1