Project Title: To Analyze Relationships and Trends of Stock Prices in Amsterdam, Frankfurt, London, Hong Kong, Japan, Singapore and New York
This project shall endeavor to establish the correlations between the stock prices in Amsterdam, Frankfurt, London, Hong Kong, Japan, Singapore and New York for the purpose of providing information that could help in making investment decisions and inferences. Apart from correlations, this project shall also focus to establish the trends of the performance of stocks in these countries by fitting necessary time series regression models. The eventual significance of the trend analysis shall be forecast of the expected performance of these stocks in the future by using the estimated parameters of our time series regression. Spectral densities of the time series data could be determined and appropriate forecasts be done using the Box-Jenkins’ Forecast Procedures.
Through time series scatter plots, we shall easily identify the outliers of the time series data, with an appropriate estimation of the standard deviation of the errors, assuming that the time series data is stationery and has serially uncorrelated errors with a zero mean. In addition to this, we could also do data regress using linear regression technique to identify the trend, identify the outliers and use the F-test technique to test whether the fit is sufficient. Mean and variance could be used to describe the central tendency and the dispersion of data.
Importance of the project
Among several benefits, the main importance of this project shall be to provide information to the investors concerning relationships, past performance and trends of stock prices the countries of our interest and what could be the expected future performance of these stocks. Certainly, this shall enable the investors to make better investment decisions in these countries.
Description of data and its limitation
We shall use a stock market dataset with 3129 values of stock prices for each country above. The number of observations is sufficiently large enough to enable application of advanced statistical methods in data analysis. The variables shall be the stock values for each country with 3129 observations, and the time variable of when they were recorded, which satisfies the characteristics of time series data. Even though this data has a major limitation of lack of variety for different particular stock prices for every country, it is sufficiently enough to meet our objective.
Conclusion
In conclusion, this project is centered in application of advanced statistical methods for stock market price analysis, in an effort to provide information of great value to the potential investors in the stock markets in Amsterdam, Frankfurt, London, Hong Kong, Japan, Singapore and New York.
Reference
Source of data:
Massey University (2012). Stock market data. Retrieved on 1st February 2013 from http://www.massey.ac.nz/~pscowper/ts/stockmarket.dat
