Testing the weak form market efficiency hypothesis for vietnam stock market from 2013 to 2017,masters thesis

Testing the Weak- Form Market Efficiency Hypothesis for Vietnam Stock Market from 2013 to 2017 Dissertation submitted in partial fulfillment of the Requirement for the MSc in Finance FINANCE DISSERTATION ON Testing the Weak- Form Market Efficiency Hypothesis for Vietnam Stock Market From 2013 to 2017 NAME OF STUDENT: Ngo Thi Minh Hoa ID No: 17047718 Intake Supervisor: Dr Tran Manh Ha 2018 Ngo Thi Minh Hoa_MSc in Finance | Testing the Weak- Form Market Efficiency Hypothesis for Vietnam Stock Market from 2013 to 2017 Executive Summary This thesis examines the effectiveness of the Vietnam Stock Exchange Daily observations are used in the period from 2013 to 2017 for the VN Index This study uses the statistical method described: unit testing, variance testing, autocorrelation test, run test of market log return to examine the effectiveness of information in the market of Vietnam stock market The results show that daily data of the VN index showed signs of failure to follow the theory of random walk The study concludes that Vietnam's stock market is not Weak- Form Market Efficiency This finding is consistent with previous studies on weak form efficiency in Vietnam Ngo Thi Minh Hoa_MSc in Finance | Testing the Weak- Form Market Efficiency Hypothesis for Vietnam Stock Market from 2013 to 2017 Table of Contents Executive Summary 2 Introduction 1.1 Purpose and contribution of research 1.2 Limitation of the study Efficient Market Hypothesis Theory 2.1 Weak form efficiency 2.2 Semi-strong form efficiency 2.3 Strong form efficiency 2.4 Testing models for EHM 2.4.1 The Fair Game Model 2.4.2 The Submartingale Model 2.4.3 The Random Walk Model 2.5 Vietnam Securities Market Overview Literature review Data 13 Methodologies for Weak Form market Efficiency Testing 15 5.1 Unit root test 15 5.2 Variance testing 16 5.3 Autocorrelation testing 17 5.4 Runs test 19 Test results analysis 21 6.1 Unit root test 21 6.2 Variance test 22 6.3 Autocorrelation test 23 6.4 Runs test…………………………………………………………………………………………………………26 Autocorrelation test (LBQ test) 25 Conclusion 27 Reference 29 Web references: 32 Appendix B: Dissertation Supervision Meeting Log 34 Ngo Thi Minh Hoa_MSc in Finance | Testing the Weak- Form Market Efficiency Hypothesis for Vietnam Stock Market from 2013 to 2017 Introduction 1.1 Purpose and contribution of research The purpose of this dissertation was to test the weak market efficiency hypothesis for the Vietnam Stock Exchange Topics include the following objectives based on the market theory of statistical validation methods in order to hypothesize that the stock market in Vietnam has a weak of information efficiency From this topic will have a broader view of the stock market in Vietnam, concluded information on Vietnam’s stock market effectively or not, thus offering solutions to the stock market Vietnam is growing 1.2 Limitation of the study Although I have tried a lot, I still have some limitations as follows The thesis only stops at the weak market test and does not carry out effective semi-strong form and strong form Thesis focuses on researching data for the whole VN index, not for individual stocks Efficient Market Hypothesis Theory Efficient Market Hypothesis (EMH) theory has important implications in practice as well as in the theory of the financial sector Economist Samuelson (1965) has said that financial economics is considered the king's jewelry on the occasion of the ritual The market theory effectively would account for half of the jewelry EMH was first introduced in 1900 by Louis Bachelier in his doctoral thesis entitled "The Theory of Speculation" and since then it has continued to improve Among the authors of the EMH doctrine are Kendall (1953) and Paul Samuelson (1965) Economists later found evidence related to EMH and by 1960 the random walk theory (a hypothesis about the relationship between past and future prices of goods - or In the 1960s, the theory of random walk continued to expand to EMH with the emergence of the EMH semi-strong form and strong form Ngo Thi Minh Hoa_MSc in Finance | Testing the Weak- Form Market Efficiency Hypothesis for Vietnam Stock Market from 2013 to 2017 Efficient market hypothesis was proposed by American scientist F Fama in 1970 Fama argues that stock prices reflect the information contained in it: weak form efficiency, semi-strong form and strong form 2.1 Weak form efficiency In the first form, a low-performing market is defined as a stock price that adequately reflects the weak information of a stock (historical information), including historical prices, volume In this weak form efficient market, the current price reflects all past earnings and all information about that stock in the market Therefore, this hypothesis means that the yields of securities, as well as other information, are not correlated In other words, investors cannot look for unusual returns with their similar information in the past This type of efficient market is often found in developing countries or emerging markets 2.2 Semi-strong form efficiency Market theory is effective at the semi-strong form, assuming stock prices reflect both past and existing public information Public information available on the market is stock price, profitability, trading volume Stock prices will adjust immediately after the information is published When market performance is at a semi-strong level, there are no predictable price changes This means that investors who make a decision based on new information will not receive any unusual profits because the current price reflects all of that publicity This type of market usually develops in developed countries around the world 2.3 Strong form efficiency The final form of the market is strong form In this market, the price of securities reflects all information relevant to it, including past and present, from the public to internal, even personal information is also reflected The strong form efficient market theory is the synthesis of both types of effective markets which are weak and semistrong forms In an efficient market for information at the level of strong form, the information provided to investors is the same, they cost the same to obtain that information and bear the same risk No investor has the right to access valuationrelated information or use it to make super-profits for themselves, because the market Ngo Thi Minh Hoa_MSc in Finance | Testing the Weak- Form Market Efficiency Hypothesis for Vietnam Stock Market from 2013 to 2017 reflects the information to investors very quickly, from the information Public information to internal information Therefore, the effective market in strong form requires the market to collect all the information 2.4 Testing models for EHM EMH is an important financial theory in the foundation of modern financial theory Fama has identified three test models for EMH: - The Fair Game Model - The Sub martingale Model - The Random Walk Model 2.4.1 The Fair Game Model The EMH theory is derived from The Fair Game theory, so The Fair Game Model is also used to test EMH in the stock market The coin toss can also be considered as the simplest example of this theory because the probability of the face and the back of the coin is the same More specifically, The Fair Game is a game in which there is no systematic difference between the actual results and the expected results before the game takes place Similarly, the stock market will be a fair game if there is no systematic discrepancy between the actual and expected returns of the stock In mathematics, it is written by the form below: ( ) ( ) : Actual observation of stock at time t + ( ) : The expected price of stock i at time t + 1, in terms of market information aggregation is : The difference between the actual price and the expected price of the stock If the efficient market according to The Fair Game Theory, there is no difference between the actual stock price and the expected stock price: ( ) [ ( )] On the other hand, if the market is efficient, the income of the stock will be similar to the stock price in the following equation: Ngo Thi Minh Hoa_MSc in Finance | Testing the Weak- Form Market Efficiency Hypothesis for Vietnam Stock Market from 2013 to 2017 - ( ); ( ) * ( )+ = : Actual observation of stock I at time t+1 : The difference between actual and expected returns of stock i ( ) : The expected return of stock I at time t + in terms of market aggregation is 2.4.2 The Submartingale Model This model comes from the Fair Game Model The new model has a small adjustment to the earnings of the stock The income of securities investors in this model will tend to increase instead of zero as the previous model Mathematically, it can be written in the form of the following equation: ( ( ) ) ( ) Although there has been a correction in earnings in the future, this model still shows the randomness in the income of securities Because even though the income of the stock will tend to increase over time or the dividend will be greater than zero But how much increase will no one predict Price appreciation is the result of market correction when new information appears on the market In the market for efficiency, all information is published very randomly; we cannot predict when there will be new information and what information Because of this important characteristic in the market efficiency, all methods of forecasting stock prices in the future are completely meaningless 2.4.3 The Random Walk Model In the stock market, the earnings per share at a given moment are interpreted as the sum of the present value of the income stream that investors will receive in the future If the market is efficient, the price per share will reflect all the information related to the situation of the company Any change in share price will affect the earnings stream of the stock Factors that affect stock prices are new information on the market However, new information or news is unpredictable So the income of the stock will also change with the new information in a way that we cannot predict Although tomorrow is almost certainly different today, it is also in a completely random way Therefore, according to EMH we have the following equation: Ngo Thi Minh Hoa_MSc in Finance | Testing the Weak- Form Market Efficiency Hypothesis for Vietnam Stock Market from 2013 to 2017 Or (*) The equation (*) is defined as the random walk for the price as well as the income of the stock Professor Fama pointed out that the random walk model was actually the expansion of The Fair Game Model If The Fair Game Model emphasizes the market's equilibrium in earnings per share, the random walk model supports the EMH theory more strongly than the theoretical methods in The Fair Game Model 2.5 Vietnam Securities Market Overview Considered the cornerstone of modern finance, Fama's The Efficient Market Hypothesis has become a common concern among investors over the years The reason is that it partly explains the internal movement of the stock market The principles of The Efficient Market Hypothesis demonstrate the nature of the random price change in the market, which is also the attribute of the market index According to this theory, an effective market is a market in which stock prices fully reflect all information related to that stock No one will be able to anticipate what the next day's stock price will be because people not know what tomorrow's information will be Stock prices in an efficient market always reflect the value of it Therefore, all acts contrary to the above rules will not be able to help investors find unusual profits for themselves (Fama, 1970) Even so far, there has not been a real-world stock market at the moment, driven by so many different causes and factors But it is a common goal of all stock markets, which is the perfect model for countries to build their own stock markets Because of its importance, EMH testing on the stock market in Vietnam will help us have a more general view of the market, know how the market is moving, develop effectively or are not So, we have timely and appropriate measures to promote the market more and more effective The stock market is the optimal channel for mobilizing capital, which plays an important role in the development of the financial market in particular and the economy in general The State Securities Commission (SSC) was established on November 28, 1996, starting with the birth of the Vietnam Securities Market The Ngo Thi Minh Hoa_MSc in Finance | Testing the Weak- Form Market Efficiency Hypothesis for Vietnam Stock Market from 2013 to 2017 stock market in Vietnam officially came into operation in July 2000 (SSC Website) Compared with other developed countries in the world, Vietnam's stock market is still very early In the first trading session on July 28, 2000, the Vietnam stock market had only two listed companies (cafef.vn) By 2005, the number of listed companies increased up to 27 enterprises Until now, Vietnam stock market has experienced more than 20 years of operation According to the latest statistics from the State Securities Commission (SSC), 2090 companies are listed on the stock exchange of Vietnam, Registered volume as of May 2018 is 125.92 billion shares Average trading value as of June 2018 was VND 14,203 billion (SSC website) Besides outstanding achievements, the stock market in Vietnam still exists many inadequacies such as the phenomenon of price manipulation, price manipulation, and violation of information disclosure On the market, there are many companies that reported losses, but soon after that, stock prices continued to rise in many sessions There were a lot of tickers going to the ceiling, down on the floor for dozens of consecutive sessions, while there was no supporting information; the issuers just explained the "supply and demand" on the market So the real question is how sensitive the reaction of securities prices to information on the stock market in Vietnam, in other words, Vietnam's stock market is effective or not and effective? In order to answer this question as well as explain some of the phenomena happening in the Vietnam securities market in the current period, I conducted a study on the subject of "testing the Weak- Form Market Efficiency Hypothesis for Vietnam Stock Market between 2013 and 2017" Literature review Over the past few decades, most empirical tests have been conducted involving each of the Random Walk Hypothesis ( RWH) or Weak Form Efficiency (WFE) implications for the stock market in both emerging and emerging markets, They produce different results Most of the early research on WFE theory is based on the RWH random step theory Many studies have examined the effectiveness of the weak market by way of income from technical analysis Alexander (1961, 1964) introduced a flexible trading strategy called "Filter Rule" This rule states that after using historical stock price data, investors should buy the stock if its price increases x%, and keep it until it decreases by x%, then he has a short sale Investors shifted from buying Ngo Thi Minh Hoa_MSc in Finance | Testing the Weak- Form Market Efficiency Hypothesis for Vietnam Stock Market from 2013 to 2017 positions when prices rose x% to selling prices Projections using this method were conducted by Fama and Blume (1966) They use a flexible business strategy (Filter rule) for a fixed period and then compare it with a buy-and-hold strategy If the profit generated by a flexible business strategy is greater than the profit generated by the passive business strategy, the stock market is considered ineffective However, in the opposite case, this could be seen as strong evidence of weak market performance in the stock market, according to Olowe (1999) The purpose of a flexible trading strategy is to allow investors to profit from any systematic trend in the price movement of securities over time Thus, Alexander (1961, 1964) and Fama and Blume (1966) concluded that there was an effective weak signal Until the late twentieth century, research by Fama and Blume (1966) was considered the leading study of business rules Another team tested the stock market's weak performance by examining the correlation of returns over a time series If random walk theory is correct then the correlation coefficient is expected to be zero Cowles and Jones (1937) compared the frequency of numerical strips to test the random walk theory using the Runs test Earlier, in Kendall's paper, he concluded that the change in stock prices was almost independent by measuring correlation coefficients Other studies by Dimson and Massavian (1998), Osborne (1959), Fama (1965), Fama and Blume (1966) also showed similar results In addition, Osborne (1959) argues that changes in the price chain are independent of the decision of the investor or that the trading is independent Fama (1970) observed that these changes in prices were mainly due to the emergence of new information Because the information is random in appearance, the movement of stock prices also follows the random step Previous studies conducted in developed countries supported WEF theory with findings of low correlation coefficients Kendall's (1953) study, for example, analyzes the changes in the stock prices of 22 US consumer securities and the UK industry's share price (1883-1934) He concluded that the price of the stock fluctuated randomly Moreover, Fama's (1965) study of the behavior of the 30 stock indexes of the Down Jones Industrial Average during the period 1957-1962 used chain testing, autocorrelation, and filtering techniques Alexander He found that the correlation coefficient was very small Therefore, it is impossible to seek extraordinary profits from business strategies Fama concluded that the Dow Jones Industrial Average was weak Solnik (1973) used 234 stock models of eight major European stock markets from 1966 to Ngo Thi Minh Hoa_MSc in Finance | 10 Testing the Weak- Form Market Efficiency Hypothesis for Vietnam Stock Market from 2013 to 2017 So the question is, in a process that is considered random, will we expect more or fewer sequences? It is obvious that too few strings will show the difference or nonrandomness of the process The statistical tool that gives us the formula for expectancy sequences is as follows: ∑ m: number of strings expected N: the total number of observations of the sample : the total number of strings in each string type With a large number of observations (N> 30), the sample distribution of the expected value (m) is treated as close to the normal distribution, and the standard deviation of the distribution is given by the formula: ∑ *∑ + ∑ } Here the hypothesis that we need to test is: the chain of observation of the daily profit of stocks and the VN index is random, ie using tail test: H0: =0 The statistic value in this case is the Z test The statistical value of Z will be compared to the value of the table as a basis for accepting or rejecting the null hypothesis Z is determined by the formula: R: number of sequences of the observed range in reality ∑ M: The number of sequences expected by chance 0.5: continuous adjustment factor This is a coefficient studied and published in 1956 by two scientists Wallis and Robert In case if the number of strings counted from reality is less than the expected value, the adjustment coefficient is continuously increased (+ 0.5) If the number of sequences counted in practice is less than the expected number of random sequences, the correction factor is continuously decreased (- 0.5) Ngo Thi Minh Hoa_MSc in Finance | 20 Testing the Weak- Form Market Efficiency Hypothesis for Vietnam Stock Market from 2013 to 2017 R < m → + 0.5 R > m → - 0.5 When the number of strings R is too large or too small will not fall within the range of the expected number of strings, the null hypothesis is rejected at this time That means the sequence of nonzero random variables that depend on observations Of course, the theory of effective market will not be accepted in the Vietnam stock market Test results analysis 6.1 Unit root test Null Hypothesis: LOG_RETURN has a unit root Exogenous: Constant Lag Length: (Automatic - based on SIC, maxlag=23) Augmented Dickey-Fuller test statistic Test critical values: 1% level 5% level 10% level t-Statistic Prob.* -38.32323 -3.434523 -2.863270 -2.567739 0.0001 t-Statistic Prob *MacKinnon (1996) one-sided p-values Augmented Dickey-Fuller Test Equation Dependent Variable: D(LOG_RETURN) Method: Least Squares Date: 09/14/18 Time: 13:40 Sample (adjusted): 1494 Included observations: 1493 after adjustments Variable Coefficient Std Error LOG_RETURN(-1) -0.992388 0.025895 C -8.52E-05 0.000219 R-squared Adjusted R-squared S.E of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) 0.496228 0.495890 0.008469 0.106947 5006.093 1468.670 0.000000 -38.32323 0.0000 -0.388540 0.6977 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat -2.48E-06 0.011928 -6.703406 -6.696295 -6.700757 1.998995 Table : Unit root test result Ngo Thi Minh Hoa_MSc in Finance | 21 Testing the Weak- Form Market Efficiency Hypothesis for Vietnam Stock Market from 2013 to 2017 From the unit root test, Augmented Dickey-Fuller test statistic is -38.32323, which too diminutive than MacKinnon tabulated value we see that the value of p-value