See the attached. This initial study was created to link the previous KSE-100 with India, Japan (Nikkei), Malaysia, and China. With the current government's initiative to turn Pakistan into a big global economy and create global integration for investment, saw the birth of the PSX (Pakistan Stock Exchange), so now the goals are beyond this initial study and the goal is to integrate with NYSE and not just South Asian / Regional Markets. After all, Pakistan has been one of the top 5 ROI based markets for the past 3 years.
The most recent document, I can't attach for obvious reasons. And it is also a huge document (over 1200 pages) on this entire Pakistani economy's globalization. But I've added some components below (some of the stuff is very dry, statistical testing and sampling) and some other documents and an image or two. Hopefully, that would set the credentials of this thread as to its not all talk, there are global organizations behind it and Morgan Stanley, and the World Bank is driving this with the Pakistanis. Check out table 1 below if you want to skip the dry stuff and see the statistical results between Pakistan's KSE (now PSX) and FTSE, NYSE, etc.
Also, Pakistan is already included in the TOP 16 fastest growing countries (BRICS + 11). See attached and below
--------------------------------------------------------------------------------------------------------------
Objective of the study
The focal objective of this study was to explore dynamic linkages between US, European,
Asian and Pakistan’s stock market which may be very useful for investors,financial
institutions and portfolio managers to utilize their capital across the border in an efficient
and smart manner.
Econometrics Methodology and Data Sources
This empirical study is based on weekly closing index of five stock markets indices of
major trading partners (US, China, India, U.K, Germany) of Pakistan. The data which
was used in this study span from 1 January, 2000 to Dec 31, 2010 (Yahoo Finance). The
return of each index is computed by the following formula.
There are different econometric techniques which are used to analyze the relationship between different time series macroeconomic variables. The study focuses to explore the association among world stock markets and Pakistan’s stock market by using the following techniques. (1) Descriptive Statistics (2) Correlation Matrix (3) Johansen Co
integration tests (4) Granger Causality test.
Johansen Julises Co integration is used in the time series data. The ideal condition to use JJ technique is that the time series variables are integrated in the same order. Stationarity of macroeconomics time series has been checked with the help of unit root test. The study used Augmented Dicky Fuller test to check the unit root in time series macroeconomicsvariables.
The ADF test checks the occurrence of stationarity in an AR (Autoregressive) equation.
The AR (1) equation is written as follows
Where Yt is the time series variables. The regression equation can be written.
(Linkages of Pakistani and Global Stock Markets Page 236)
Whereis the first difference operator. This equation may be used to estimate and
testing for a unit root is equivalent to testing null hypothesis Most of the financial time series variables are stationarity at first difference. If two series are integrated of at same order, there may have a linear combination that may be
stationary without differencing.There are two methods of co integration testing. Engle Granger (1987) test and Maximum likelihood base Johansen (1988; 1991) and Johansen - Juselius (1990) tests.
Both tests check the long run association among the macroeconomic variables. The likelihood ratio
of Johansen Juselius test the number of co integration vectors in VAR system. Eigenvalue test check the null hypothesis that there is at most r numbers of co integration equation in VAR. the maximum Eigen value statistic is given by 1
---------------------------------------------------------------------------------------------------------------
Table 1: Descriptive Statistics
KSE NYSE BSE SSE FTSE DAX
Mean 0.410909 0.039088 0.325421 0.166558 -0.017812 0.035978
Median 0.727541 0.244188 0.705154 0.000000 0.156427 0.339018
Maximum 13.64966 12.89389 14.07764 14.96379 13.40915 16.11623
Minimum -18.20683 -19.53480 -15.95417 -13.84131 -21.04692 -21.60969
Std. Dev. 3.738035 2.761628 3.552023 3.625219 2.674297 3.547779
Skewness -0.721553 -0.660061 -0.419698 0.203003 -0.664536 -0.267582
Kurtosis 5.877906 9.792691 5.165853 4.809623 12.34745 7.030426
Jarque-Bera 236.2333 1091.342 122.9726 78.39371 2031.676 376.7631
Probability 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000
Sum 224.7670 21.38096 178.0055 91.10697 -9.743155 19.67980
Sum Sq. Dev. 7629.205 4164.116 6888.812 7175.650 3904.917 6872.356
Observations 547 547 547 547 547 547
---------------------------------------------------------------------------------------------------------------
Another test is based on the “trace statistic” which tests the null hypothesis of r cointegrating vectors against the alternative hypothesis of r. The test is checked by following trace value” (Kerry Patterson).“The important thing i
s to apply JJ test is the selection lag length of VAR model. The lag length is selected on the basis of AIC” (Akaike Information Criterion).
Analyzing the long run relationship among the variables, then the system of the VAR model should be converting into error correction term to account for the short run dynamics of variables from their long run equilibrium.
According to the “Granger theorem” if two time series variables are co integrated, it means that there must be at least one direction causality exist between the time series variables. JJ test and ECM only capture the long and short run dynamics of equilibrium.
Hence, the chronological Granger causality between the macroeconomics variables may be analyzed with the joint F test. The advantage of the Granger causality test is that it can check the lead and lag relationship between the variables within the sample period. This type of exercise is also made by the variance decomposition and impulse response test.
Variance decomposition and impulse response analysis accurately measure the shock of the values of one variable in a given period which arising from the same variables as well as other variables in previous periods. The equation of the impulse response function and variance decomposition is written.
View attachment 298723
