quantitive techniques in business
DESCRIPTION
Final projectQuantitive Techniques in BusinessTRANSCRIPT
Quantitive Techniques in Business
Final projectQuantitive Techniques in BusinessA report Submitted to The Department of Management and Administrative Sciences, University of Gujrat in Partial Fulfillment of the Requirements For the Degree of Master in Business Administration
Submitted By:Muhammad Jahanzaib (11022720-028)MBA 4th (3.5Year) Session (2011-2015)
Name of Program Coordinator: Mr. Mirza Rizwan Sajid
UNIVERSITY OF GUJRAT Date of Submission 27/06/2014
Table of ContentS. NoDescription
1Introduction
2Type of analysis
3Univariate test
4Bivariate test
5Association test
6Correlation test
7Regression Analysis
8Simple regression model
9Multiple regression model
10Assumption of regression model
11Normality
12Linearity
13Homocedascity
14Autocorrelation
15Multicolinerity
16Means compare test
17Parameter test
18Simple t test
19Independence t test
20Compare t test
21ANOVA
22Non parameter test
23Factor Analysis
24Explanatory Factor Analysis
25Confirmatory Factor Analysis
Introduction
The project is about the application of the tools and techniques studied in the subject quantitative technique in business. During my semester, I have studied number of techniques and their practical application in the software. I have applied all the tools and techniques studied in this course using SPSS. This work proves very helpful in research work as well as for data analysis. In this project, I used 5 variables out of which 2 variables were qualitative (Nominal) and remaining 3 were quantitative variables. Tools and techniques which are practiced in this project include: univariate Analysis, Bivariate Analysis, Multivariate analysis, histogram, pie charts, bar chart, regression analysis including simple regression and multiple regression models, parametric and non-parametric tests, mean comparison and factor analysis with interpretation.
Types of analysis
There are three types of analysis:
Univariate Analysis Bivariate Analysis Multivariate Analysis
Univariate Analysis
Quantitative variables
My quantitative variables include
Serial NoVariables
1Debt ratio
2Bankruptcy Risk
3Liquidity Risk
1. Debt RatioDescriptive Statistics
NMinimumMaximumMeanStd. Deviation
Debts Ratio50.211.09.5332.20868
Valid N (list wise)50
Histogram
Interpretation:The Debt ratio is Quantitative variable and for quantitative variables we make histogram. According to results the average Debt Ratio of the respondents is 0.5332 and minimum value is 0.21 and maximum value is 1.09. While standard deviation is 0.20868, Subject vise interpretationOn the basis of given sample information we can conclude that firm average debt use more than 50% of its capital and minimum debt are use 0.21 and some firm use debt more than of its whole equity.
2. Bankruptcy RiskDescriptive Statistics
NMinimumMaximumMeanStd. Deviation
Bankruptcy Risk503.356.765.1300.85697
Valid N (list wise)50
Histogram
Interpretation:
The Bankruptcy risk is Quantitative variable and for quantitative variables we make histogram. According to results the average value is 5.13. Minimum value is 3.35 and maximum value is 6.76, while standard deviation is 0.8569.Subject vise interpretationOn the basis of given sample information we can conclude that firms are bear bankruptcy risk average 5.13 that result show the all the firms must face liquidity risk.
3. Liquidity Risk Descriptive Statistics
NMinimumMaximumMeanStd. Deviation
Liquidity Risk50.303.481.5554.81707
Valid N (list wise)50
Histogram
Interpretation:The Liquidity risk is Quantitative variable and for quantitative variables we make histogram. According to results the average value is 1.5554, and Minimum value is 0.30 and maximum value is 3.48, and the standard deviation is 0.81707.
Subject vise interpretationOn the basis of given sample information we can conclude that firm are average liquidity is 1.55 firms must maintain liquidity close to 1 because firm are liabilities are easily pay at any time.
Qualitative Variables
Followings are my qualitative variables:
Serial NoVariables
1Debt Strategies
2 Profit Loss
1. Debt Strategies
Firms are debts strategies implement/ Non implement
FrequencyPercentValid PercentCumulative Percent
ValidImplement2550.050.050.0
non implement2550.050.0100.0
Total50100.0100.0
Bar chat
Interpretation:The Debt strategies are qualitative variable and in case of qualitative variable we make bar chart or pie chart. According to given results out of 50 respondents 50% of the respondents are implement the debt strategies on firms and 50% of respondents are not implement the debt strategies on firms.
Subject vise interpretationOn the basis of given sample information we can conclude that half firms uses debt strategies and half firms use not implement the debt strategies.
2. Profit & Loss
Company Earn profit Or suffering Loss
FrequencyPercentValid PercentCumulative Percent
Validprofit2856.056.056.0
Loss2244.044.0100.0
Total50100.0100.0
Interpretation:This is qualitative variable and in case of qualitative variable we make bar chart or pie chart. According to our results 44% respondents have their bearing loss and remaining 56% of respondents enjoying a profit.
Subject vise interpretationOn the basis of given sample information we can conclude that 44 percent firms suffering losses and 56 percent earn profit.
Bivariate Analysis
Quantitative Variables
1. Debt Ratio & Bankruptcy Risk2. Debt Ratio & Liquidity Risk
Debt Ratio
One-Sample Kolmogorov-Smirnov Test
Debts_Ratio
N50
Normal ParametersaMean.5332
Std. Deviation.20868
Most Extreme DifferencesAbsolute.103
Positive.103
Negative-.088
Kolmogorov-Smirnov Z.730
Asymp. Sig. (2-tailed).661
a. Test distribution is Normal.
Interpretation
1. Hypothesis Formulation Ho: Data is Normal.H1: Data is not normal.2. Level of Significance: = 0.053. Test-statistic: One sample t-test4. Critical Region: p > 5. Calculations: p = 0.6616. Conclusion: To check the normality assumption we applied KS Test. Since P-Value is 0.661 which is Greater than 0.05. So we will reject our Alternative hypothesis. So our data is normal.
Bankruptcy Risk
One-Sample Kolmogorov-Smirnov Test
Bankruptcy_Risk
N50
Normal ParametersaMean5.1300
Std. Deviation.85697
Most Extreme DifferencesAbsolute.139
Positive.070
Negative-.139
Kolmogorov-Smirnov Z.982
Asymp. Sig. (2-tailed).290
a. Test distribution is Normal.
Interpretation
1. Hypothesis Formulation Ho: Data is Normal.H1: Data is not normal.2. Level of Significance: = 0.053. Test-statistic: One sample t-test4. Critical Region: p > 5. Calculations: p = 0.2906. Conclusion: To check the normality assumption we applied KS Test. Since P-Value is 0.290 which is Greater than 0.05. So we will reject our Alternative hypothesis. So our data is normal.
Association test
firms are debts strategies implement/ Non implement * company Earn profit Or suffering Loss Cross tabulation
Count company Earn profit or LossTotal
ProfitLoss
firms are debts strategies implement/ Non implementimplement18725
non implement101525
Total282250
Interpretation1. There is generating hypothesis: H0: There is no association between debts strategies and profit. H1: There is association between debts strategies and profits.2. Level of Significance: = 0.053. Test-statistic: Chi-square test4. Critical Region: p < 5. Calculations: p = 0.023 and chi square = 5.1956. Conclusion: On the basis of given sample information we conclude that there is an association between debt strategies and profit. These firms are implement the debt strategies are earn profit or good performed, but on the other hand these firms are not implement strategies that are more chance to suffering loss or low performed in market.
Subject vise interpretationOn the basis of given sample information we can conclude that association between that firms use dept strategies are earn profit and that are not apply these are suffering loss.
Correlation
Correlations
Debts RatioBankruptcy Risk
Spearman's rhoDebts RatioCorrelation Coefficient1.000.506**
Sig. (2-tailed)..000
N5050
Bankruptcy RiskCorrelation Coefficient.506**1.000
Sig. (2-tailed).000.
N5050
**. Correlation is significant at the 0.01 level (2-tailed).
Interpretation1. There is generating hypothesis: H0: There is no correlation between Debt ratio and bankruptcy risk. H1: There is correlation between Debt ratio and bankruptcy risk.2. Level of Significance: = 0.053. Test-statistic: Chi-square test4. Critical Region: p < 5. Calculations: p = 0.000 and Pearson Test = 0.5376. Conclusion: Data is normal distribution thus used the Pearson Test of correlation the Probability of significance is 0.000 less than selected P-.value thus null hypothesis is rejected. On the basis of given sample information we conclude that there is a highly strong correlation between debt Ratio and bankruptcy Risk.
Subject vise interpretationOn the basis of given sample information we can conclude that result shows that if the bankruptcy risk increase debt ratio also increases if bankruptcy risk decrease debt ratio also decrease there is highly strong positive relationship between bankruptcy risk and debt ratio.
Regression Model
1. Simple Linear Regression2. Multiple Linear Regression
Simple Linear Regression
Model Summary
ModelRR SquareAdjusted R SquareStd. Error of the Estimate
1.537a.289.274.17783
a. Predictors: (Constant), Bankruptcy Risk
Interpretation:According to model summary results of simple linear regression model shows that R-square value is 0.289 which is less significant. This value says that the simple regression (Bankruptcy risk) explains only 28.9% of the variation in dependent variable (Debt Ratio).
1. Hypothesis Formulation:Ho: Model is not fit for prediction. H1: Model is fit for prediction.2. Level of Significance: = 0.053. Test-statistic: ANOVA4. Critical Region: p < 5. Calculations: p = 0.000
ANOVAb
ModelSum of SquaresdfMean SquareFSig.
1Regression.6161.61619.477.000a
Residual1.51848.032
Total2.13449
a. Predictors: (Constant), Bankruptcy Risk
b. Dependent Variable: Debts Ratio
Conclusion: According to given results we in a position to reject Ho. So we have to accept Alternative Hypothesis that model is fit for perdition
Coefficients
ModelUnstandardized CoefficientsStandardized CoefficientsTSig.
BStd. ErrorBeta
1(Constant).138.154-.895.375
Bankruptcy Risk.131.030.5374.413.000
a. Dependent Variable: Debts Ratio
Interpretation:According to our table the values: = 0.1381 = 0.131So Our Simple linear equation: Y = (0.138) + (0.131) X Subject vise interpretationOn the basis of given sample information we can conclude that independent variable is more effect to dependent variable if bankruptcy risk is increase than also increase debt ratio. If bankruptcy risk decrease than debt ratio also decreased.
Multiple Linear Regression Model
Regression Model:Debts Ratio = + 1 (Bankruptcy Risk) + 2 (Liquidity Risk)
Model Summary
ModelRR SquareAdjusted R SquareStd. Error of the Estimate
1.627a.393.367.16598
a. Predictors: (Constant), Liquidity Risk, Bankruptcy Risk
Interpretation:According to model summary results of multiple linear regression model shows that R-square value is 0.393 which is significant. This value says that in the multiple regressions model (independent variables) explains 39.3% of the variation in dependent variable (Debt ratio).
1. Hypothesis Formulation:Ho: Model is not fit for prediction. H1: Model is fit for prediction.2. Level of Significance: = 0.053. Test-statistic: ANOVA4. Critical Region: p < 5. Calculations: p = 0.000
ANOVAb
ModelSum of SquaresdfMean SquareFSig.
1Regression.8392.41915.227.000a
Residual1.29547.028
Total2.13449
a. Predictors: (Constant), Liquidity Risk, Bankruptcy Risk
b. Dependent Variable: Debts Ratio
6. Conclusion: According to given results we are in a position to reject Ho. And alternative hypothesis accepts so we have to accept that model is fit for perdition.
Coefficients
ModelUnstandardized CoefficientsStandardized CoefficientstSig.
BStd. ErrorBeta
1(Constant).229.1931.187.241
Bankruptcy Risk.088.032.3602.782.008
Liquidity Risk-.094.033-.369-2.846.007
a. Dependent Variable: Debts Ratio
Interpretation:According to our table the values:
= .2201 = 0.0882 = -.094
Debts Ratio = + 1 (Bankruptcy Risk) + 2 (Liquidity Risk)
Debts Ratio = .229 + .088 (Bankruptcy Risk) + -0.94 (Liquidity Risk)
Subject vise interpretationOn the basis of given sample information we can conclude that both independent variables are effect to dependent variable if bankruptcy risk and liquidity risk increase debt ratio increase if decrease debt ratio decrease.
Assumptions of linear regression
1. Normality:
One-Sample Kolmogorov-Smirnov Test
Debts_Ratio
N50
Normal ParametersMean.5332
Std. Deviation.20868
Most Extreme DifferencesAbsolute.103
Positive.103
Negative-.088
Kolmogorov-Smirnov Z.730
Asymp. Sig. (2-tailed).661
a. Test distribution is Normal.
Interpretation
1. Hypothesis Formulation Ho: Data is Normal.H1: Data is not normal.2. Level of Significance: = 0.053. Test-statistic: One sample t-test4. Critical Region: p > 5. Calculations: p = 0.6616. Conclusion: To check the normality assumption we applied KS Test. Since P-Value is 0.661 which is Greater than 0.05. So we will reject our Alternative hypothesis. So our data is normal.
Histogram
Interpretation:It is also clear from the histogram that data is not normal. For showing normality it should be of bell shaped. This is more on the left side. This histogram is positively skewed.
P.P Plot for Saving (Dependent Variable)
Interpretation:The results in p-p plot shows that data is not normal.
2. Linearity
Scatter DiagramDiagram of Debt ratio, Bankruptcy risk and liquidity risk
Interpretation:We can see from the scatter diagram that no variable is linear. The relationship between Debt ratio and Bankruptcy risk is not leaner and relationship between Debt ratio and Liquidity risk is almost leaner.
Scatter DiagramDiagram of debt ratio, debt strategies and profit & loss
Interpretation:We can see from the scatter diagram that no variable is linear. The relationship between Debt ratio and Debt strategies is not leaner and relationship between Debt ratio and Profit &Loss is not leaner.
3. Homocedascity
Independent Samples Test
Levines Test for Equality of Variancest-test for Equality of Means
FSig.tdfSig. (2-tailed)
Debts RatioEqual variances assumed2.456.1242.09648.041
Equal variances not assumed2.16847.934.035
Bankruptcy RiskEqual variances assumed.189.6662.69548.010
Equal variances not assumed2.73347.203.009
Liquidity RiskEqual variances assumed3.325.074-1.82048.075
Equal variances not assumed-1.75837.888.087
Interpretation:
On the basis of given sample information to conclude that we are apply a Levines test for Equality of variances all the variables significance level set 0.05, but our results are debt ratio is 0.124, bankruptcy risk is 0.666 and liquidity risk is 0.074 which is all p values are insignificance. Thus to conclude that in our sample data are not homocedascity that is hetrocedascity.
4. Auto Correlation
Model Summaryb
ModelRR SquareAdjusted R SquareStd. Error of the EstimateDurbin-Watson
1.627a.393.367.165980.917
a. Predictors: (Constant), Liquidity Risk, Bankruptcy Riskb. Dependent Variable: Debts Ratio
InterpretationApplying the independence test the value of Durbin-Watson is 0.917 thus in variables problem of auto correlation occurred. This value is not lying between 1.5 -- 2.5. So there is serious effect of auto-correlation.
ANOVAs
ModelSum of SquaresdfMean SquareFSig.
1Regression.8392.41915.227.000a
Residual1.29547.028
Total2.13449
a. Predictors: (Constant), Liquidity Risk, Bankruptcy Risk
b. Dependent Variable: Debts Ratio
5. No Multicorenarity
Coefficients
ModelUnstandardized CoefficientsStandardized CoefficientstSig.Collinearity Statistics
BStd. ErrorBetaToleranceVIF
1(Constant).229.1931.187.241
Bankruptcy Risk.088.032.3602.782.008.7701.299E0
Liquidity Risk-.094.033-.369-2.846.007.7701.299E0
a. Dependent Variable: Debts Ratio
Interpretation
In the coefficient table we have to check the tolerance and VIF under colinearity Statistics. Here the T value should be greater than 0.2 and VIF values less than 5 are considered favorable. In table we can see that T value of all variables is more than 0.2 more over VIF value is less than 5 which are favorable values.
Collinearity Diagnosticsa
ModelDimensionEigenvalueCondition IndexVariance Proportions
(Constant)Bankruptcy RiskLiquidity Risk
112.8101.000.00.00.02
2.1813.936.01.03.60
3.00918.122.99.97.38
a. Dependent Variable: Debts Ratio
InterpretationIn collinearity diagnostics the condition index values less than 30 shows no big issue of multicollinarity. In our results the value condition index 18.122 thus there is no multicollonarity in our debt.
Mean Comparison
S. NoParametric TestsNon-Parametric Tests
1One Sample T-TestSign test
2Independent Sample test Independent sample /Mann Whitney U-test
3Paired Sample test Wilcoxen Sign rank test
4ANOVA Kruskal- wallindependent test
Parametric Test
1. One Sample T-Test
Debt Ratio
For the application of parametric tests I am supposing that the variable Debt ratio have normal data. 1. Hypothesis Ho: Average Debt ratio =0.50H1: Average Debt ratio 0.502. Level of Significance: = 0.053. Test-statistic: One sample t-test4. Critical Region: p < 5. Calculations: p = 0.000
One-Sample Statistics
NMeanStd. DeviationStd. Error Mean
Debts Ratio50.5332.20868.02951
One-Sample Test
Test Value = 0
TDfSig. (2-tailed)Mean Difference95% Confidence Interval of the Difference
LowerUpper
Debts_Ratio18.06649.000.53316.4739.5925
6. Conclusion: On the basis of given data we are not in a position to accept Ho. Since P-Value is 0.000 which is less than 0.05. So we will reject our null hypothesis. So average family saving is Average Debt ratio 0.50
2. Independent Sample Test
1. Hypothesis Ho: Data is Normal.H1: Data is not normal. 2. Level of Significance: = 0.053. Test-statistic: One sample t-test 4. Critical Region: p < 5. Calculations: p = 0.124 & p = 0.196
One-Sample Kolmogorov-Smirnov Test
GroupCGPA
MaleN30
Normal ParametersaMean3.1357
Std. Deviation.14738
Most Extreme DifferencesAbsolute.271
Positive.165
Negative-.271
Kolmogorov-Smirnov Z1.485
Asymp. Sig. (2-tailed).124
MaleN20
Normal ParametersaMean3.5840
Std. Deviation.23266
Most Extreme DifferencesAbsolute.241
Positive.241
Negative-.223
Kolmogorov-Smirnov Z1.078
Asymp. Sig. (2-tailed).196
a. Test distribution is Normal.
6. Conclusion: On the basis of given data we are not in a position to reject Null Hypothesis Since P-Value is 0.124 and 0.196 which is more than 0.05. So we will reject our null hypothesis. So our data is normal
1. Hypothesis Ho: Ho= f = mH1: Ho= f m2. Level of Significance: = 0.053. Test-statistic: Independent Sample Test4. Critical Region: p < 5. Calculations: p = 0.00022Group Statistics
GroupNMeanStd. DeviationStd. Error Mean
CGPAmale303.1357.14738.02691
Male203.5840.23266.05202
Independent Samples Test
Levene's Test for Equality of Variancest-test for Equality of Means
FSig.tdfSig. (2-tailed)Mean DifferenceStd. Error Difference95% Confidence Interval of the Difference
Lower
CGPAEqual variances assumed1.595E1.00022 -8.3553748.000-4.48333E-15.36581E-2-.55622
Equal variances not assumed-7.654492.916E1.000-4.48333E-15.85713E-2-.56810
6. Conclusion:On the basis of given sample we are rejected to Ho. Since P-Value is 0.00022 which is less Than 0.05 so we can say that CGPA Male and females are not same.
3 Paired Sample Test1. Hypothesis Ho: Data is Normal.H1: Data is not normal. 2. Level of Significance: = 0.053. Test-statistic: One sample t-test 4. Critical Region: p < 5. Calculations: p = .039
One-Sample Kolmogorov-Smirnov Test
difference
N50
Normal ParametersaMean.1098
Std. Deviation.20903
Most Extreme DifferencesAbsolute.199
Positive.199
Negative-.141
Kolmogorov-Smirnov Z1.405
Asymp. Sig. (2-tailed).039
a. Test distribution is Normal.
6. Conclusion: According to results we are rejected to H0. So our Data is not normal. But we assume data are normal for checking the parametric test (paired sample test).
Paired sample test Hypothesis Ho: Ho= B = A (CGPA before and CGPA are equal after gaudiness)H1: H1= B E (CGPA before and CGPA are not equal after gaudiness)Level of Significance: = 0.05Test-statistic: Paired Sample StatisticsCritical Region: p < Calculations: p = 0.001
Paired Samples Statistics
MeanNStd. DeviationStd. Error Mean
Pair 1CGPA3.315050.28822.04076
After CGPA3.424850.29287.04142
Paired Samples Test
Paired DifferencestdfSig. (2-tailed)
MeanStd. DeviationStd. Error Mean95% Confidence Interval of the Difference
LowerUpper
Pair 1CGPA After CGPA-1.09800E-1.20903.02956-.16920-.05040-3.714E049.001
6. Conclusion:According to results the p-value is less alpha value. So we are rejected to H0 and we have to accept Null hypothesis CGPA before and after gaudiness are not equal. ANOVA1. Hypothesis Ho: Data is Normal.H1: Data is not normal. 2. Level of Significance: = 0.053. Test-statistic: One sample t-test 4. Critical Region: p < 5. Calculations: p = .019 & .109 & .220
One-Sample Kolmogorov-Smirnov Test
Group_AnovaCGPA
Lower familyN16
Normal ParametersaMean3.0756
Std. Deviation.14213
Most Extreme DifferencesAbsolute.381
Positive.191
Negative-.381
Kolmogorov-Smirnov Z1.522
Asymp. Sig. (2-tailed).019
Middle FamilyN18
Normal ParametersaMean3.2633
Std. Deviation.06371
Most Extreme DifferencesAbsolute.284
Positive.284
Negative-.218
Kolmogorov-Smirnov Z1.206
Asymp. Sig. (2-tailed).109
Upper FamilyN15
Normal ParametersaMean3.6667
Std. Deviation.20931
Most Extreme DifferencesAbsolute.271
Positive.254
Negative-.271
Kolmogorov-Smirnov Z1.051
Asymp. Sig. (2-tailed).220
a. Test distribution is Normal.
6. Conclusion:According to results only business man variable data is not normal. We supposed that our data is Normal. So now we apply ANOVA Test on it
ANOVA:1. Hypothesis Ho: Ho= S = B = E (CGPA of lower class student, middle and upper is Equal)H1: H1= S B E(CGPA of lower class student, middle and upper is not equal) 2. Level of Significance: = 0.053. Test-statistic: One Way ANOVA4. Critical Region: p < 5. Calculations: p = 0.000
ANOVA
CGPA
Sum of SquaresdfMean SquareFSig.
Between Groups2.41212.41269.812.000
Within Groups1.65848.035
Total4.07049
6. Conclusion: On the basis of given data we are no rejected Ho. Since P-Value is 0.000 which is less than 0.05. So we can say that CGPA of lower class student, middle and upper are not equal.
Non-Parametric Tests
Mann-Whitney Test1. Hypothesis Ho: Ho= m = f (CGPA of male and female are equal)H1: H1= m f (CGPA of male and female are not equal)2. Level of Significance: = 0.053. Test-statistic: Mann-Whitney Test4. Critical Region: p < 5. Calculations: p = 0.000
Ranks
GroupNMean RankSum of Ranks
CGPAMale3015.50465.00
Male2040.50810.00
Total50
Test Statistics
CGPA
Mann-Whitney U.000
Wilcoxon W465.000
Z-5.995
Asymp. Sig. (2-tailed).000
a. Grouping Variable: Group
6. Conclusion: According to the results the P-value is less than the alpha value and on the basis of this p-value we are rejected the null hypothesis. So we have to accept that CGPA of male and females are not equal.
Wilcoxon Signed Ranks Test1. Hypothesis Ho: Ho= B = A (Saving before and saving are equal after increase in salary)H1: H1= B A (Saving before and saving are not equal after increase in salary)2. Level of Significance: = 0.053. Test-statistic: Wilcoxon Signed Ranks Test4. Critical Region p < 5. Calculations: p = 0.00
Ranks
NMean RankSum of Ranks
Group CGPANegative Ranks50a25.501275.00
Positive Ranks0b.00.00
Ties0c
Total50
a. Group < CGPA
b. Group > CGPA
c. Group = CGPA
Test Statistics
Group CGPA
Z-6.170a
Asymp. Sig. (2-tailed).000
a. Based on positive ranks.
b. Wilcoxon Signed Ranks Test
6. Conclusion:As per the given data the results shows that the p-value is less than alpha value which is 0.000 < 0.05. So we are not in a position to accept Ho. So we have to accept that CGPA before and after Gaudiness not equal.
FACTOR ANALYSIS
1. Explanatory Factor Analysis2. Confirmatory Factor Analysis
Explanatory Factor Analysis
1. Hypothesis Ho: Ho= Data is suitable for Factor Analysis. H1: H1= Data is not suitable for Factor Analysis.2. Level of Significance: = 0.053. Test-statistic: KMO and Bartlett's Test4. Critical Region: p < 5. Calculations: p = 0.00
KMO and Bartlett's Test
Kaiser-Meyer-Olkin Measure of Sampling Adequacy..693
Bartlett's Test of SphericityApprox. Chi-Square35.911
df3
Sig..000
6. Conclusion:According to the KMO value given in results 0.0.693 we can say that sample is 69.3% that is Almost sufficient that is close to 70%. On the basis of given data we can say that p-value is less than alpha value so null hypothesis is rejected. So our data is not suitable for factor analysis.
Communalities
InitialExtraction
Debts Ratio1.000.714
Bankruptcy Risk1.000.661
Liquidity Risk1.000.665
Extraction Method: Principal Component Analysis.
Interpretation:As per the given data the initial values of all the variables are 1 showing that each variable is playing 100% role. But in extraction, the values greater than 0.6 are favorable. Variables having extraction values above 60% are favorable in extracting information and these variables are playing important role in factor analysis.
Total Variance Explained
ComponentInitial EigenvaluesExtraction Sums of Squared Loadings
Total% of VarianceCumulative %Total% of VarianceCumulative %
12.04067.98867.9882.04067.98867.988
2.52017.33785.326
3.44014.674100.000
Extraction Method: Principal Component Analysis.
InterpretationOn the basis of given data one variable have total value more than 1 so these variables should be included into the factor. This variable explains 67.98% variation. So we have to accept this variable for factor analysis.
Interpretation
The Screen Plot shows that all the points above 1 should be included into factor. There is one point above 1.
Component Matrix
Component
1
Debts Ratio.845
Bankruptcy Risk.813
Liquidity Risk-.815
Extraction Method: Principal Component Analysis.
a. 1 component extracted.
InterpretationOn the basis of given sample information only one variable have been generated, this is given below:Factor 1: Debt ratio, Bankruptcy risk and liquidity risk.
Rotation competent Matrix Only one component extracted thus the solution are not generated.
Muhammad Jahanziab11022720-028Page 37