Chapter 8

Analysis of Primary Data

147
 CHAPTER 8

ANALYSIS OF PRIMARY DATA

32 questionnaire forms received by way of primary data collection was loaded

to the SPSS software for the Initial analysis. Overall, there were 18 criteria,

which were having an impact on the performance of the Open Ended Equity

Schemes of Mutual Fund. However, to come out with a Discriminant Analysis

output with all the 18 Criteria would lead to a very lengthy model. To counter

the same, the Factor Analysis was used for the data reduction. For the data

reduction the feedback was taken from Fund Manager / Assistant Fund

Managers / Backup Fund Managers / Research Analysts.

Factor Analysis Results

The Factor Analysis was applied for the identification of the core factors

affecting the performance of Mutual Fund Schemes (Open Ended Equity

Schemes). This technique was considered appropriate as it requires no pre-

existing of functional relationships and is a well known for data reduction. It is

used to reduce large number of variables into a few numbers of core factors.

148
Test Adequacy of Sample

The Kaiser-Meyer-Olkin is the measure of sampling adequacy, which varies

between 0 and 1. The values closer to 1 are better and the value of 0.6 is the

suggested minimum. The Bartlett's Test of Sphericity is the test for null

hypothesis that the correlation matrix has an identity matrix. Taking this into

consideration, these tests provide the minimum standard to proceed for

Factor Analysis.

Test hypothesis regarding interrelationship between the variables.

Null Hypothesis H0 : There is no statistically significant interrelationship

between variables affecting the performance of Mutual Fund Scheme.

Alternate Hypothesis H1 : There may be a statistically significant

interrelationship between variables affecting the performance of Mutual Fund

Scheme.

149
SPSS Output :

KMO and Bartlett's Test

Kaiser-Meyer-Olkin Measure of Sampling

.702
Adequacy.

Bartlett's Test of Approx. Chi-Square

543.556
Sphericity

Df 153

Sig. .000

Normally, 0 < KMO < 1

If KMO > 0.5, the sample is adequate.

Here, KMO = 0.702 which indicates that the sample is adequate and we may

proceed with the Factor Analysis.

Bartlett’s Test of Sphericity

Taking a 95% level of Significance, α = 0.05

The p-value (Sig.) of .000 < 0.05, therefore the Factor Analysis is valid

150
As p < α, we therefore reject the null hypothesis H0 and accept the alternate

hypothesis (H1) that there may be statistically significant interrelationship

between variable.

The Kaiser-Meyer Olkin (KMO) and Bartlett's Test measure of sampling

adequacy was used to examine the appropriateness of Factor Analysis. The

approximate of Chi-square is 543.556 with 153 degrees of freedom, which is

significant at 0.05 Level of significance. The KMO statistic of 0.702 is also

large (greater than 0.50). Hence Factor Analysis is considered as an

appropriate technique for further analysis of the data.

Eigen values (Select those components with Eigen Values >= 1)

The initial components are the numbers of the variables used in the Factor

Analysis. However, not all the 18 variables will be retained. In the present

research only the 4 factors will be extracted by combining the relevant

variables. The Eigen values are the variances of the factors. The total

column contains the Eigenvalue. The first factor will always account for the

most variance and hence have the highest Eigen values. The next factor will

account for as much of the left over variance as it can and the same will

continue till the last factor. The percentage of variance represents the percent

of total variance accounted by each factor and the cumulative percentage

gives the cumulative percentage of variance account by the present and the

151
preceeding factors. In the present research the first 4 factors explain 75.95%

of variance.

The rotation sums of the squared loading represent the distribution of the

variance after the varimax rotation with Kaiser Normalisation. The varimax

rotation tries to maximize the variance of each of the factor.

152
 SPSS Output :

Eigen Values – Total Variance Explained

Total Variance Explained

Compo Extraction Sums of Squared Rotation Sums of Squared
nent Initial Eigenvalues Loadings Loadings
% of % of
Varian Cumulati Varianc Cumulati % of Cumulati
Total ce ve % Total e ve % Total Variance ve %
1 7.894 43.857 43.857 7.894 43.857 43.857 5.300 29.445 29.445
2 2.883 16.016 59.874 2.883 16.016 59.874 4.215 23.414 52.860
3 1.848 10.269 70.143 1.848 10.269 70.143 2.642 14.679 67.539
4 1.046 5.810 75.952 1.046 5.810 75.952 1.514 8.413 75.952
5 .910 5.056 81.009
6 .883 4.905 85.914
7 .721 4.004 89.918
8 .422 2.344 92.263
9 .292 1.623 93.886
10 .261 1.453 95.339
11 .240 1.336 96.675
12 .165 .919 97.594
13 .150 .835 98.429
14 .135 .750 99.179
15 7.644
.425 99.604
E-02
16 5.330
.296 99.900
E-02
17 1.325 7.362E
99.974
E-02 -02
18 4.762 2.646E
100.000
E-03 -02
Extraction Method: Principal Component Analysis.

On the basis of Varimax Rotation with Kaiser Normalisation, 4 factors have

been extracted. Each factor is constituted of all those variables that have

factor loadings greater that 0.5. 18 variables were clubbed into 4 factors. 4

153
factors were extracted from the 18 variables used in the study. These 4

extracted factors explained 75.95% of the variability the performance of Open

Ended Equity Schemes of Mutual Funds. This explains over three-fourth of

the variability.

Scree Plot

The scree plot graphs the Eigenvalue against the each factor. We can see

from the graph that after factor 4 there is a sharp change in the curvature of

the scree plot. This shows that after factor 4 the total variance accounts for

smaller and smaller amounts.

154
SPSS Output :

Scree Plot
10

4
Eigenvalue

0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Component Number

Identification of the Core Factors

The Rotated Factor Matrix represents the rotated factor loadings, which are

the correlations between the variables and the factors. The factor column

represents the rotated factors that have been extracted out of the total factor.

These are the core factors, which have been used as the final factor after data

reduction. According to the grouping of the factors, each group of factors is

named which will represent the grouped factor and represent the factors.

155
SPSS Output :

Rotated Component Matrix(a)

Component
1 2 3 4
Returns of 3 Year Period -.188 -.241 .740 -.256
Mean Return 8.439E-
.307 .755 .267
02
Standard Deviation of the
-.017 .187 .711 -.382
Returns
Coefficient of variation of the
.255 .638 .345 .380
returns
NAV percentage change in
.138 -.288 .755 .159
the 3 Year Period
Geometric mean of the
Excess Return of the 7.878E-02 .380 -.042 .834
Benchmark
Value at Risk (VAR) .343 .579 .439 .104
Sharpe Index 1.133E-
.166 .750 .292
02
Modigliani Measure .490 .656 -.112 .117
Information Ratio .319 .537 -.129 .328
Beta Coefficient β .159 .721 -.032 -.235
Treynor Index .487 .693 -.147 .143
Jensen Alpha α Coefficient .150 .758 -.203 .197
Treynor & Mazuy's α 7.801E-
.943 .257 .105
Coefficient 02
Treynor & Mazuy's γ 9.578E-
.908 .235 .143
Coefficient 02
Henriksson & Merton's α 8.225E- 6.900E-
.951 .196
Coefficient 02 02
Henriksson & Merton's γ 5.697E-
.936 .208 .118
Coefficient 02
Treynor & Black Appraisal
.903 .293 -.017 -.012
Ratio
Extraction Method: Principal Component Analysis.
Rotation Method: Varimax with Kaiser Normalization.

a Rotation converged in 9 iterations.

156
The above matrix gives the correlation of the variables with each of the

extracted factors. Usually, each of the variables is highly loaded in one factor

and less loaded towards the other factors. To identify the variables, included

in each factor, the variable with the value maximum in each row is selected to

be part of the respective factor. The values have been highligtened in each of

the rows to group the 18 variables into 4 core factors.

Thus, after rotation, Factor 1 accounts for 29.44% of the variance; Factor 2

accounts for 23.42% of the variance; Factor 3 accounts for 14.67% of the

variance; Factor 4 accounts for 8.42% of the variance. All the 4 factors

together explain for 75.95% of the variance in performance of Open Ended

Equity Scheme.

157
Name of the four core factors

The variables that have been included into each core factor have been named

as under: -

Table 8 : Name of the four core factors

Factor Variables Included Name of the

Factor
1 • Treynor & Mazuy's α Coefficient Stock
• Treynor & Mazuy's γ Coefficient Selection &
• Henriksson & Merton's α Coefficient Timing
• Henriksson & Merton's γ Coefficient
• Treynor & Black Appraisal Ratio
2 • Coefficient of Variations of the Returns Risk
• Value At Risk (VAR) Management
• Sharpe Index
• Modigliani Measure
• Information Ratio
• Beta Coefficient (β)
• Treynor Index
• Jensen Alpha α Coefficient
3 • Returns of 3 Year Period Existing
• Mean Return Returns of
• Standard Deviation of the Returns the Schemes
• NAV percentage change in the 3 Year (At
Period Beginning)
4 • Geometric Mean of the Excess Return Excess
over the Benchmark Return over
Benchmark
(At
Beginning)

158
Conclusions of Factor Analysis

The Factor Analysis has thus identified 4 core factors that affect the

performance of the Open Ended Equity Schemes. They can be categorized

as under: -

1 Stock Selection and Timing

2 Risk Management

3 Existing Returns of the Schemes (At Beginning)

4 Excess Return over the Benchmark (At Beginning)

The above factors have been discussed in detail as under: -

Factor 1 – Stock Selection and Timing

This factor suggests the market timing skill and security selection ability of the

Fund Managers. The first factor explains 29.44 % of the variability on the

performance of the Open Ended Equity Scheme. It is necessary for the Fund

Managers to have a superior Market Timing Skill and the Security selection

ability to out performance the benchmark. Similarly, this is the core factor,

which contributes to the performance of an Open Ended Equity Scheme.

159
Factor 2 – Risk Management

The second factor relates to the market risk management. This factor

explains 23.42% of the variability on the performance of the Open Ended

Equity Scheme. Equity Schemes are more risky and depends on the stock

markets for its performance. Thus the Fund Managers have to obviously take

care of the market risk management.

Factor 3 – Existing Returns of the Scheme (At Beginning)

The third factor characterizes the existing returns of the Open Ended Equity

Scheme. It is always said that those schemes, which perform better, tend to

perform better in the future. This factor explains 14.67% of the variability on

the performance of the Open Ended Equity Scheme.

This result is in conformity with that of Hendricks et al, 1994, Brown and

Goetzman, 1995 who have found evidence supporting the idea that past

performance is related to future performance.

Factor 4 – Excess Return over the Benchmark (At Beginning)

The fourth factor characterizes the excess return over the benchmark. This

factor explains 8.42% of the variability on the performance of the Open Ended

160
Equity Scheme. An Open Ended Equity Scheme has to outperform the

Benchmark and generate higher returns than the Benchmark so that its

performance is better.

A total of 4 core factors extracted from the 18 variables, explains 75.95% of

the variability of the performance of the Open Ended Equity Schemes.

Table 9 : Identification of Variables related factors in Variables Selection

Factor Name Variables Included Factor

Loading
Stock • Treynor & Mazuy's α Coefficient .972
Selection & • Treynor & Mazuy's γ Coefficient .909
Timing • Henriksson & Merton's α Coefficient .954
• Henriksson & Merton's γ Coefficient
• Treynor & Black Appraisal Ratio .937

.901
Risk • Coefficient of Variations of the
Management Returns .735
• Value At Risk (VAR)
• Sharpe Index .656
• Modigliani Measure .676
• Information Ratio .697
• Beta Coefficient (β) .514
• Treynor Index .601
.760
• Jensen Alpha α Coefficient
.677
Existing • Returns of 3 Year Period .707
Returns of • Mean Return .743
the Schemes • Standard Deviation of the Returns .687
(At • NAV percentage change in the 3 .697
Beginning) Year Period
Excess • Geometric Mean of the Excess .848
Return over Return over the Benchmark
Benchmark
(At
Beginning)

161
Based on above factor loadings, the variables within each extracted factor

were identified for further Discriminant Analysis. This data was framed into a

Discriminant Model to generate a predictive evaluation for the performance of

the Open Ended Scheme based on the selected coefficients / variables

values.

Analysis of Part B of the questionnaire viz Factor that Lower Risk

The Part B of the questionnaire was on the factors that lower risk vis-à-vis the

systematic risk and the unsystematic risk. The same scale of 1 to 6 was

considered to rate these two factors of the total risk of a portfolio.

It is useful in balancing portfolio to distinguish between two sources of risk:

market risk or systematic risk on the one hand and appraisal or insurable risk

on the other. In general it is not correct to assume that optimal balancing

leads either to negligible levels of appraisal risk or to negligible level of market

risk (Treynor Jack & Black Fisher).

The total risk in a portfolio can be expressed as under: -

Total Risk = Systematic Risk + Unsystematic Risk

162
Systematic Risk

The systematic risk is also referred to as the market risk and cannot be

diversified. It refers to the movement of the whole market. Even if the Fund

Manager has a perfectly diversified portfolio, there is a risk, which cannot be

avoided, and this is known as the systemic risk. The systemic risk is not the

same for all the securities or portfolio.

Unsystematic Risk

Unlike the systematic risk, the unsystematic risk is a specific risk, specific to

the company or industry that is inherent in each investment. The

unsystematic risk can be reduced through proper diversification. This is also

referred to as the non-market risk. This risk is specific to a particular security

and is associated with factors like the business and financial risk.

There was 18.75% mortality from the Part A of the questionnaire to the Part B

of the questionnaire. There was 81.25% of feedback from the respondent.

As the response rate was high, it was felt that we may proceed with the

descriptive analysis of the data.

This data was again framed on the SPSS software for further analysis and

interpretation. The descriptive statistics was used to analyse the data

163
collected in the part B of the questionnaire in the primary data. There were

two attributes in this data viz factor that lower risk viz the Systematic Risk and

the Non Systematic Risk.

SPSS Output : Descriptive Analysis

Statistics

Systematic Risk Non Systematic Risk

N Valid 26 26
Missing 6 6
Mean 2.9615 2.9615
Median 3.0000 3.0000
Mode 2.00 1.00
Sum 77.00 77.00

The N value gives the Valid and Missing number of the valid observations for

the variables. In the present observations out of the 32 feedbacks, the

respondent gave 26 feedback and 6 of the respondent did not give the

feedback. The mean value viz the arithmetic mean across the observation for

the systematic and non-systematic means is the most widely used central

tendency. The mean for both these criteria has come out to be exactly

2.9615, as the central tendency, which connotes that both the systematic risk

and the non-systematic risk are equally important factor after lower risk, which

has to be considered for the portfolio.

164
Going by the mode weightage, however it was felt at the systematic risk is a

marginal important factor that has to be considered to lower risk of the

portfolio.

Analysis of Part 'C' of the questionnaire viz Other Useful Inputs or

Remarks

The questionnaire was a semi structured questionnaire wherein the part 'C' of

the questionnaire was on any other useful inputs or remarks that the Fund

Managers would like to contribute towards the present research. There were

few feedback, which was related to the present research. The feedback have

been capsulated as under:

• The returns of 1 year, 2 years, 5 years (are important for performance).

• Other qualitative factor are also important in terms of Portfolio

concentration – large cap – small cap – mix etc

• The list was exhaustive (For the data collection)

• Primarily performance is relative to peer group is the key.

• Don’t diversify too much across Sectors.

• High Portfolio trading volume indicates lack of clarity.

• What is largely seen is returns of the funds vis-à-vis Benchmark

depends on the market psychi – sometimes preference for large cap /

165
 mid cap / small cap will influence returns or sector preference would

influence it more than any other logical / mathematical measure.

• Consistency in the funds performance should also be looked into. The

fund which outperforms both in a rising & falling market should be

preferred.

All the above input on performance was considered as valid in context to the

present research. Largely, it has been seen that the existing returns have

been considered as the influential factor for the future performance of the

scheme. The returns over a benchmark / peer group was the factor which

was considered as the primarily performance. In the present research, the

benchmark was taken as the BSE Sensex, which again endorses the views

expressed by the Fund Managers. Consistency in the funds performance was

also a factor to be looked into, the mean return however gives the mean of the

3 years and therefore these views were also taken care in the present

research.

166