Analysis and Interpretation Using IBM SPSS

Get Complete Project Material File(s) Now! »

Analysis and Interpretation

The purpose of this chapter is to evaluate and analyze the empirical findings from responses to the questionnaire through processing of raw data using IBM SPSS software, and interpretation of these data based on the main arguments, theories, and empirical findings included in the theoretical framework in chapter 3.
First, table 2 and figures 2 and 3 exhibit an initial processing of raw data to generate average quantitative values for Q5 and Q6 that represent respondents’ degree of loss aversion, and Q7 and Q8 that cover respondents’ degree of herding behavior. This is followed by analysis and interpretation of the data represented in the table and figures. Second, the generated average values together with the other collected data are included in and processed by IBM SPSS using the multiple regression analysis tool, following the Descriptive Statistics, Model Summary, ANOVA, and Coefficients tables. These tables provide processed data that are further interpreted using measures of model fit, T-test, F-test, the critical value approach, and the P-value approach. Finally, further interpretation and analysis of the results from the statistical testing methods is provided based on the authors’ both implementation of the main arguments, theories, and empirical findings within the theoretical framework, and consideration of other relevant aspects.

Descriptive Analysis

Referring to table 2 and figures 3 and 4, they include average frequencies of respondents’ level of loss aversion and herding behavior. As the table and figures illustrate, the initial evaluation of collected data exhibit strong loss aversion observed among the respondents. Most respondents score either 4 or 5 for Q5 and Q6 which specifically address their financial loss aversion in a hypothetical scenario in which they are exposed to a certain financial gain and loss. This strong level of loss aversion observed among the respondents is perfectly in line with and support findings from the existing empirical studies conducted by scholars such as Kahneman and Tversky (1979, 1992), Thaler and Johnson (1990), Kahneman et al., (1991), Thaler and Rabin (2001), and Beckman et al., (2011).
To expand, the results of the conducted questionnaire exhibits a strong level of financial loss aversion among the respondents similar to the loss aversion described by Kahneman and Tversky’s (1979, 1992) prospect theory. Referring to the underlying assumptions of PT, the collected data show that while facing financial uncertainty and risk most of the respondents would rather choose the option that bring them an absolute or certain financial gain or enable them to avoid an absolute or certain financial loss. This advocates the assumption of PT in which most of The Impact of Loss Aversion Bias on Herding Behavior of Young Swedish Retail Investors individuals are bounded by loss aversion bias i.e., they find the pain of losing a specific amount of financial wealth much stronger than the joy of gaining the same amount (Kahneman & Tversky, 1979, 1992).
Regarding the empirical findings for Q7 and Q8 which measure respondents’ herding degree, the average degree of herding behavior is more equally spread. Referring to the frequencies of average herding levels, it is realized that scores of 4, 3, and 2 are most observed, with equal frequencies been observed for scores of 1 and 5. This frequency distribution shapes a histogram for average degree of herding behavior which is slightly skewed toward stronger herding scores. Consequently, this right skewed graph may indicate that although the findings do not illustrate any strong herding pattern, they may picture the respondents’ positive tendency toward financial herding in hypothetical scenarios that include financial uncertainty and risk.
In other words, the slight positive skew may relatively support herding behavior theories by scholars such as Benarjee (1992), Prechter (2001), Prechter and Parker (2007), Bikhchandani et al., (1992), Fernandez et al., (2011), and Salganik et al., (2006). Considering the real life normal financial market conditions prevailing while this research being conducted, an initial evaluation of the empirical findings may also support the empirical studies by scholars such as Galariotis et al., (2014). According to Galariotis et al., (2014), no or low degree of herding behavior is observed when market conditions are normal or there is not any significant market downturn. The shape of the average herding histogram may also support findings by Chiang and Zheng (2010) that a lower degree of herding is often observed among developed markets, Sweden in this case, in comparison to emerging markets. However, Litimi’s (2017) findings indicate that herding is often observed among retail investors who have significant part of their investments in certain industry sectors such technology and telecom. Thus, his findings contradict the pattern observed in figure 4.
Therefore, the fact that respondents’ portfolio allocation choices are not included in the questionnaire and its probable impact on the results must be acknowledged. Finally, an intuitive comparison of the shape of figures 3 and 4 may reveal that the frequencies within these figures do not follow a similar pattern. To elaborate, strong average degree of loss aversion observed in figure 3 do not follow a strong average degree of herding behavior in figure 4. Instead, the frequencies in figure 4 follow a pattern independent to those observed in figure 3.

READ Financial and non-financial performance

1. Introduction
2. Problem and Purpose
3. Literature Study and Frame of Reference
3.1 Traditional Finance .
3.2 Behavioral Finance.
3.3 Hypothesis Formulation
4. Methodology
4.1 Research Paradigm
4.2 Target Population and Sample Group .
4.3 Method of Data Collection
4.4 Method of Data Analysi
4.5 Assumptions
5. Results and Empirical Findings
6. Analysis and Interpretation
6.1 Descriptive Analysis
6.2 Analysis and Interpretation Using IBM SPSS
6.3 Loss Aversion and Control Variables .
6.4 Other Factors
7. Conclusion
8. Discussion
8.1 Theoretical Contribution
8.2 Practical Implications
8.3 Limitations
8.4 Suggestions for Future Studies
9. Reference list
10. Appendices