خلاصة:
In financial distress studies selection of variable is commonly basedon the success of variables in variable sets employed in earlierbankruptcy studies, suggestions in the literature or an accompanyingdata reduction in a large set of variables. If seemingly different variablesets exhibit a strong relationship then heterogeneous variable setscapture common information. Canonical correlation analysis appropriatelyexamines the relationship between two sets of measured variables.The main purpose of the present study was to illustrate the value ofvariable deletion strategies in canonical correlation analysis for moreparsimonious to capture common information. In research contents, thelaw of parsimony states that the fewer variables used to explain asituation, the more probable that the explanation will be closer to reality.Therefore, as variable sets become more parsimonious there are greaterprobabilities that the results of the analysis will be replicable. Todetermine the common information between variable sets in financialdistress studies, the study selected two specific bankruptcy models:Altman, the most famous model, and Deakin, the biggest model. Theresults indicated that as the number of variables increase, the probableeffect of these sources of error variation on the canonical correlationincreases. Therefore, the goal of a variable deletion strategy is toestimate as much variance with the smallest variable set possible. In thisstudy the goal was achieved by removing the three variables in variablesets employed in selected bankruptcy studies.
ملخص الجهاز:
The main purpose of the present study was to illustrate the value of variable deletion strategies in canonical correlation analysis for more parsimonious to capture common information.
To determine the common information between variable sets in financial distress studies, the study selected two specific bankruptcy models: Altman, the most famous model, and Deakin, the biggest model.
Bankruptcy Studies, Variable Deletion Strategies Canonical Correlation Analysis Introduction Numerous corporate failure prediction models have so far been developed, based on various modeling techniques and financial ratios.
To determine the common information between variable sets, The study selects two bankruptcy studies and employs canonical correlation analysis to examine the relationships that exist between two variable sets and then to illustrate the value of applying the law of parsimony to canonical correlation analysis (CCA) solutions.
Thorndike (1978) stated that “as the number of variables increase, the probable effect of these sources of error variation on the canonical correlation increases” Thompson (1991) showed that CCA subsumes all other parametric methods including t-tests, ANOVA, regression, MANOVA and discriminate analysis.
3. Statistical method Canonical correlation analysis determines the extent of the relationship between two variable sets with redundancy coefficients.
Table 1: canonical correlation analysis to examine the relationships that exist between two variable sets FUNCTION3 FUNCTION2 FUNCTION1 Variablestatistic rs2 rs2 rs2 Altman model 22.
The results of canonical analysis in table 1 indicate that the pooled redundancy coefficient of the Altman set with respect to the Deakin variants is 0.