MSIN 0112 GROUP ASSIGNMENT Group C Do people really exhibit stock name biases Introduction Background and Literature Review Is a stock name and code number critical to stock investors More specifically, would investors buy stocks because of the priority preference to the name or code number of this stock Those questions are important for understanding the biases on stock name and code when investors making investment decisions. To make optimal investments, investors spend tremendous amount of money and efforts to acquire stock-related information. In 2016, Xing et al. pointed out that there is a positive relationship between stock ticker symbol and the firm value. The features of stock names were linking to the reflection of investors’ feelings, which are indeed affecting the assessment of firm values. Before conducting a comprehensive research on the financial information of specific stocks, investor’s first impression to a stock, which is commonly presented by the name and the trading code, may have substantial influence on their investment choices. This psychological predisposition could be thought as derived from individual’s preferences for certain numbers or words. Knewtson and Sias (2010) pointed out that personal preferences comfort people in a way that generating favourable reaction under certain circumstances, which in turn had impact on investor’s choices of stock selection. Furthermore, factors influencing the preferences to numbers are worth to pay attention to. According to Yang (2011), the use of superstitious numbers in pricing and the exploitation of superstition in retail sales was more than a cultural phenomenon. People have personal superstitious numbers around the world. In most of the western countries, number 13 is considered as a symbol of bad luck. Some Italians are superstitious about Friday the 17th because rearranging the Roman numeral XVII creates the world VIXI – translated from Latin meaning “my life is over”. And in China, number 4 sounds like the Chinese word for death, thus some buildings in China skip number 4 in symbolizing floors. There are two main contributions to the literature. Firstly, we extended this topic to Chinese stock market, which is a rapidly emerging and animated market in the world, having growing impact on the international level of the world economy. Secondly, numbers rather than letters are used to stand for the symbols of stocks, from where we go further to prove that the results of Xing et al. (2016) could be applied to symbols with numbers as well. Research Question The aim of our research is to test whether people exhibit stock name biases in China’s stock market. Two Hypotheses 1) In China’s stock market, people would invest less in the stocks with symbols they like. 2) In China’s stock market, people would invest less in the stocks whose symbols contain special numbers. Method Participants We summarize the statistics of our participants referring to Pickering (2017)’s structure. Our sample includes 20 observations. The ages of our participants range from 20 to 53 years. The mean of the ages is 31.95, and the standard deviation is 11.4546. 40% of our participants are males, and 60% are females. Table 1 Observations n = 20 Age Mean 31.95 Standard Deviation 11.4546 Min 20 Max 53 Gender Male 40% Female 60% Materials To research on the topic, we decide to use survey as our research method. The reasons are as follows. Firstly, compared to experimental designs, non experimental designs (such as case studies and correlational studies) are more natural. Additionally, within the non experimental designs, we cannot apply statistical analysis on case studies, and the results from case studies cannot be generalized on population because of limited samples (1, 2 or a few people). However, the aim of our research is to generalize population parameters through the statistical analysis of sample statistics. Thus, we discuss the topic through correlational studies. To be specific, we design a questionnaire and do surveys to collect the data we need for correlational studies. Our questionnaire strictly follows the principles of survey. For example, to protect the participants’ personal information, our questionnaire does not collect the participant’s name, address, telephone number, etc. One example is that all of the participants are the people we know, including the group members’ friends and relatives, and the age is between 18 and 65 years old. The questionnaire contains five questions. Through asking for the minimum questions for the study, we respect the participants’ time and privacy, and reduce the difficulty of analysing the data. Even though the number of questions is limited, question types are diversified. For instance, we design open-ended questions when asking about age, birthday and lucky number of the participant. Besides, fixed questions are used to divide gender and mark the how participants like symbols. Design and procedure First, ask participants open-ended questions about their background. The aim of this part is to obtain descriptive statistics of our sample, e.g. 1) Please write down your age. “I am 24 years old.” Besides the use of describing samples, we can use the results of this question to eliminate the samples not meeting the requirement (age between 18 and 65) as well. 2) Please choose your gender. “I prefer not to tell you my gender.” Considering the diversity of personality and the willingness of the participants to state genders, we add another two options “Others” and “Prefer not to tell” in addition to “Male” and “Female”. For hypothesis 1: In China’s stock market, people would invest less in the stocks with symbols they like. 1) IV: Likeability of symbols. Ask participants questions to measure the IV, e.g. We randomly choose 50 stock symbols from the stock list of China’s market in Yahoo Finance. The question is design to measure the scale. The participants are asked to use the number 1-5 to rate the extent to which they like a certain stock symbol. To be specific, 5 = strongly like, 4 = like, 3 = neither like nor dislike, 2 = dislike and 1 = strongly dislike. One possible answer might be “I strongly dislike the symbol 600449”. 2) DV: Prices of stocks. According to the research method of Xing et al. (2016), the impact was measured by stock’s Tobin’s Q. Thus, we consider the prices of stocks as our DV. The prices are acquired from Yahoo Finance. 3) Calculate the total score for each symbol, and the log number of stock price. 4) Calculate the correlation between the total scores and the log prices. For hypothesis 2: In China
’s stock market, people would invest less in the stocks whose symbols contain special numbers (traditional/personal lucky numbers and birthdays). 1) IV: The extent to which the symbols are special to the participants. Ask participants questions to measure the IV, e.g. We ask the participants for their birthday, which is considered as special numbers to them. Possible answer: “0420”. In addition, the participants’ personal lucky numbers are asked for. Possible answer: “7”. Traditionally, Chinese consider “6” and “8” as lucky numbers. We allocate one score to the symbol when it matches one principle mentioned above. 2) DV: Prices of stocks, which is the same as the DV mentioned in the first hypothesis. 3) Calculate the total score for each symbol, and the log number of stock price. 4) Calculate the correlation between the total scores and the log prices. Results Hypothesis 1 Table 2 Variable ln Price Score for likeability 0.0679*** (6.17) Observations 20 R-squared 0.4422 Table 2 reports the regression result of the impact of likeability of symbols on the prices of stocks, and DV is stock price in logarithmic form. The regression result shows the estimated coefficient of likeability of symbols is statistically significant at the level of 1% (p = 0.000), thus we reject the null hypothesis. Additionally, the coefficient is positive, indicating the increase of likeability is conducive to raise stock price. On average, when the likeability increases 1 score, the stock price would increases 0.06%. Our result is in accordance with the results from Xing et al. (2016), and goes further to prove the conclusion also applies to symbols with numbers rather than letters. Hypothesis 2 Table 3 Variable ln Price Score for special degree -0.0021 (-0.08) Observations 20 R-squared 0.0001 Table 3 reports the result of special degree of symbols on prices of stocks, and DV here is log stock price. The result shows the coefficient of special degree is not statistically significant (p = 0.933), thus the null hypothesis cannot be rejected. There might be several reasons. First, the special degree of a symbol has no significant relationship with stock price. The investment behaviours of the investors in China’s stock market are mostly influenced by other factors, such as the past prices of the stock, or the performance of the firm. Additionally, the principles we choose (traditional/personal lucky numbers and birthdays) might not fully represent the special degree of symbols. Conclusion General Discussion The existence of stock name biases misleads investors’ investment choices on stocks, and followed by influences on stock values. As the developing impact of China’s stock market, this study make a survey on 20 participants to test whether investors exhibit stock name biases in this specific stock market. Two hypotheses are conducted to demonstrate our research objective. The result for the first hypothesis shows that, in China’s stock market, people would invest more in the stocks with symbols they prefer. However, the result of the second hypothesis fails to show a significant relationship between investment choices on stocks and the superstitious numbers in stock symbols. Therefore, to conclude, Chinese stock market shows a stock name biases to some extent. Limitations and Suggestions for Further Research Only 20 individuals participated in the survey for this topic, and most of the participants are friends and families of the authors, some of whom even have no experience in entering the stock market. It is not sufficient to fit in the model and delivery an accurate and convincing result with small database provided by limited sample size. As Yang (2011) pointed out that in Chinese market, people pay extra money on products with price tags fitting their superstitious preferences. Yet our model fails to show a significant relation between stock price and individual’s preference on superstitious numbers in the second hypothesis. Additionally, because of the narrow scope, our sample cannot stand for the average investors in China. Besides limited sample data available, our sample obviously contains biases, such as geographic bias. It does not include participants from enough provinces and areas in China, as regional and cultural differences are highly noticeable across the country, thus the superstitious level on symbols of participants also differs. Further researchers could conduct survey on large sample to obtain abundant and sufficient database to support the model. Big data modelling with appropriate statistical method is believed to deliver a more accurate and authoritative result. Furthermore, regional differences should be taken into consideration. Moreover, in order to display a more in-depth research in this topic, whether participants having experiences in stock market impact the result is worth to investigate in. Word count: 1778 Reference Knewtson, H. S., & Sias, R. W. (2010). Why Susie owns Starbucks: The name letter effect in security selection. Journal of Business Research, 63(12), 1324-1327. Pickering, R. M. (2017). Describing the participants in a study. Age and Ageing, 46(4), 576– 581. Xing, X., Anderson, R. I., & Hu, Y. (2016). What sa name worth The impact of a likeable stock ticker symbol on firm value. Journal of Financial Markets, 31, 63-80. Yang, Z. (2011). “Lucky” numbers, unlucky consumers. The Journal of Socio-Economics, 40(5), 692-699. Appendix Questionnaire 1) [Open-ended] Please write down your age. _____ 2) [Fixed] Please choose your gender. _____ a. Male b. Female c. Others d. Prefer not to tell 3) [Open-ended] Please write down your birthday (yydd, e.g. 0405). _____ 4) [Fixed] What is your lucky number between 0 and 9 _____ 5) [Fixed] [Likert scale] There are 50 symbols of stocks in China’s market listed below. Please use the number 1-5 to mark how you like each symbol. (5 = strongly like, 4 = like, 3 = neither like nor dislike, 2 = dislike, 1 = strongly dislike) Symbol Score Symbol Score 000038 002270 601968 601138 603557 600276 002472 000002 300625 601998 002297 603288 002562 600104 300497 600663 002628 601088 000908 000651 600571 000333 002148 600000 600449 601166 002847 000858 000919 600900 600117 601328 603579 600028 002101 601628 600106 600036 002060 601988 603916 600519 000779 601857 603681 601318 300220 601398 002892 601939 Results from Stata (screen shots) 1) Descriptive statistics 2) Regression for hypothesis 1 3) Regression for hypothesis 2