PREDICTIBILITY OF RETURNS ON MICROSOFT STOCK PRICE Ziqian Zhu MSc Mathematical Finance Queen Mary University of London 13/04/2015 The aim of this article is to investigate the dynamic behavior of daily returns of Microsoft stock prices from 21/03/2005 to 21/03/2015, which are downloaded from the NASDAQ website and include 2514 observations Microsoft is an American multinational corporation that develops, manufactures, licenses, supports and sells computer software, consumer electronics and personal computers and services. Standard and Poor’s and Moody’s have both given a AAA rating to Microsoft, whose assets were valued at $41 billion on November 14, 2014 as compared to only $8.5 billion in unsecured debt and …show more content…
This means that the null hypothesis about normality can be rejected and daily stock prices are not normally distributed. Histogram 1 II. Stationarity of Microsoft daily stock prices and daily returns As the stationarity of stock prices can strongly influences its behavior and properties and the use of non-stationary data can lead to spurious regressions, thus test of stationarity is more than necessary in research of daily stock prices. I applied Augmented Dickey-Fuller test here to check for unit root. The results for the daily stock prices appear as in Table 1. Table 1 The value of the test statistic and the relevant critical values given the type of test equation and sample size, are given in the first panel of the output above. Clearly, the test statistic is -1.261047 that is not more negative than the critical value, so the null hypothesis of a unit root in the stock price series cannot be rejected. Since one of the independent variables in this regression is non-stationary, it is not appropriate to examine the coefficient standard errors or their t-ratios in the test regression. Now repeat all of the test for first difference of the daily stock prices. The output appear as in the following Table 2. Table 2 The test statistic is -34.10940 which is more negative than the critical value and hence
15 In testing the hypotheses: H0 β1 ’ 0: vs. H1: β 1 ≠ 0 , the following statistics are available: n = 10, b0 = 1.8, b1 = 2.45, and Sb1= 1.20. The value of the test statistic is:
To avoid spurious results, unit root tests using Augmented Dickey-Fuller (ADF) (Dickey and Fuller, 1981) and Philips-Perron (PP) are performed to determine the time-series properties of the variables employed in the analysis. Two or more variables are said to be co-integrated when they exhibit long run equilibrium (relationship) if they share common trend(s).Therefore Auto-regressive distributed lag bounds approach (ARDL) is used to test it. The choice is based on several
The three-year SAIC stock price data and its corresponding SSE index are obtained from finance.yahoo.com, as it provides dividend-adjusted closing prices. The two data are ordered in time in Excel (Sort Ascending). It is found that 46 SAIC daily stock prices are missing due to suspension of trading, therefore; 46 corresponding SSE daily index are removed in order to match up dates on the two data series.
Microsoft Corporation is an American global technology firm develops, manufactures, and sells computer software, electronics, and personal computers. The firm has its headquarters in Redmond, Washington. Bill Gates founded the corporation in 1975 to develop and sell necessary interpreters; it rose with time to become the dominant player in the computer operating system operating system it had its initial public offer (IPO) IN 1986. Its fast growth in the industry has enabled it to acquire Skype technologies in 2011 at the cost of $8.5 billion. The firm has plans to make more acquisitions this year. The profit margin of the company lies at US$16.79 billion.
This report is issued in order to inform the public about Microsoft Corporation. We analyzed the profitability and liquidity of this company. In addition, we were able to provide recommendations for investments or credits in Microsoft for the best interest of the public.
For d1, t-statistic=- 2.5334, t-statistic > t-critical. Thus we reject Ho and d1 is significant.
In the first week, the stock maintained its stability through the first 4 days but ended up taking a small dip on the last day of the week, which dropped the stock price from $118.69 to $116.40. In the second week, the stock dropped when the market opened back up then stabilized for the rest of the week. The stock price went from $116.40 to $112.88. In the third week, the stock increased considerably from the previous week when the market opened back up and was stable. The price was stable throughout the week until the last day of the week when the price of the stock ended up fallen. The stock price went from $112.88 to $112.40. In the fourth week, the stock market opened back up. The price of the stock had went down but increased dramatically and ended up rising back to its original price. The stock price went from $112.40 to $114.16. In the fifth week, the stock increased considerably. When the market opened back up, the stock price was slow and steady on the first day and then started increased throughout the rest of the week almost reaching back to the purchasing price. The stock price went from $114.16 to $117.20.
Microsoft is a highly diversified company. Its technologically-related products span from software to music players to game consoles to web browsers to search engines to phones. However, its flagship product, the product which has been the primary driver of its profits has been Microsoft Windows, the ubiquitous operating system that runs on virtually every computer in the world. Windows has been deemed so critical that even Microsoft's competitor Apple was effectively forced by market pressures to allow its Macs to run Windows, in an effort to boost sales. "As astounding as Apple's success has been, it hasn't put a dent in the Microsoft Office monopoly. [Current CEO] Ballmer and company still profit on every Macbook running Word, Excel and PowerPoint" (Greg 2012).But while Microsoft continues to make its highly profitable Windows products (despite industry criticism about its user features); it has struggled to diversify in its many critical areas, most notable in its music, phones, and Internet service.
Using the statistical tables in your textbook, find the values of the appropriate test statistic in the following two situations:
First of all, variables should be given in log levels in order to alleviate the problem of serial correlation and the elasticity of the coefficients. The results of ADF unit root test in levels concludes that all three variables - seasonally
As indicated by the case study S&P 500 index was use as a measure of the total return for the stock market. Our standard deviation of the total return was used as a one measure of the risk of an individual stock. Also betas for individual stocks are determined by simple linear regression. The variables were: total return for the stock as the dependent variable and independent variable is the total return for the stock. Since the descriptive statistics were a lot, only the necessary data was selected (below table.)
Testing the stationary properties of time series is a very important exercise as the use of stationary time series data in the Classical Linear Regression Model will result in inflated results. The results are likely to be inconsistent and with a low Durbin Watson (DW) statistic. Several methods can be employed to test whether the time series variables are stationary , these includes residual plot but this paper will employ the Augmented Dickey Fuller (ADF) to test the existence of a unit root. Conclusion of stationarity is going to be considered at 1% and 5% level of significance only. Any probability of each variable below the two values will be considered stationary. If the model fails to meet the stationary requirement, we will use the differencing method to make our model stationary.
In order to test for a long-term relationship in the Sri Lankan economic model, we use the Engle-Granger residual-based approach with all the variables and test for cointegration.In this case, we test the stationarity ofβ_1 x_1t+β_2 x_2t+⋯β_n x_nt=e_t. This approach is based on the Chapter 6(pp.343) in “Applied econometric time series” (Enders, 2015).
It should be emphasized, however, that because of inability of the DF-GLS to capture the possibility of a structural break, the power of the test is likely to decrease with an undetected structural break in the series, thereby providing misleading results. For completeness, therefore, we investigate unit roots in the presence of a structural break using the Perron-Vogelsang (PV) test. The results of the PV test are summarized in Panel B of Table 1. As seen in the DF-GLS test, the null cannot (can) be rejected for all the levels (first differences) of the variables, confirming that all the three variables are I(1) series. Hence, it is certain that the underlying series are apparently all I(1) processes even after taking into account a structural break in the series,
The critical value according to Narayan (2005) (Case III: Unrestricted intercept and on trend) No trend, K = 5, (***), (**), (*) denotes Significant at 1%, 5% and 10% respectively.